While I basically agree with you, in that programming is a task best done well or not at all, it's a common mistake to think that everyone who programs should be a career programmer. It must be possible to teach the necessary programming skills (and they are necessary) to college students in engineering, science, and a few other fields. It's a little bit like mathematics except that many students don't start learning it early enough. If one doesn't start learning programming before college, then it's possibly too late to really learn enough in college. I think these days really successful students primarily learn it outside of formal classes.
My take on the topic of Fortran is that it is not important to learn in school. I've written a lot of Fortran, but with a strong background in C it was not hard to learn when it was necessary for me to. Even in experimental physics, Fortran is used increasingly rarely so most students would not have a need to use it.
I think that learning one programming language in school is not nearly enough. Students should learn basic procedural programming concepts, something about data structures, object oriented design, and good practices in some scripting language to get comfortable with hashes and regular expressions. Python is great for a lot of this, but it's probably better to establish more low-level concepts first, including pointers. Maybe learning C and then Python could be a reasonable, minimalistic approach? C++ could actually be avoided since the syntax and rules are so complex. (I'm not talking about comp-sci majors who should certainly learn C++.) Perl is awesome of course, but again the syntax is complex and in the context of a college course that just takes time away from learning the essential concepts. Java is possible but doesn't really get either low-level or high-level enough for my tastes.
Whatever a student learns, they should have enough breadth, conceptual knowledge, and practice so that whatever language is required for a given project would not be too hard to learn on the job.
I think you're right about the acoustics problem being non-trivial. There are many factors to consider in the design of instruments. I'd guess that the first problem to solve is how one can reproduce the notes called for in Bach's score. The shape of the mouthpiece, bore, and bell will affect the pitches that it resonates at. Real brass instruments play notes in a kind of harmonic sequence, and notes get closer together as you go up the scale. Sounds like this instrument is a little like a french horn in that it's about the same length (about the same as a trombone too) but is played high (maybe 3 octaves or more?) above the fundamental. But the exact shape of the instrument is what dictates the frequencies of these modes of vibration. Probably even the stiffness of the brass makes a small difference. And beyond the notes it can make, each resonant mode will have a kind of width, or Q value, that relates to how easy it is to bend the pitch to something in a 12-tone scale. A related problem is working out the timbre or general sound of the instrument, also a function of the shape. I don't know how the designers would have any reference for that.
I don't know how they went about modeling the instrument. I guess the basic approach would be to consider a geometry and then sweep through frequencies, looking for resonances. But how would you simulate the buzzing of lips in a mouthpiece? If they just put a microphone up to a mouthpiece of some shape and had a musician play different pitches, that still wouldn't account for the feedback or back-pressure of the rest of the instrument.
While I know enough physics and acoustics to be arrogant, I still would hardly know where to start on such a project. I'm really curious to learn about the details of what they did.
SDL supports Linux, Windows, Windows CE, BeOS, MacOS, Mac OS X, FreeBSD, NetBSD, OpenBSD, BSD/OS, Solaris, IRIX, and QNX. The code contains support for AmigaOS, Dreamcast, Atari, AIX, OSF/Tru64, RISC OS, SymbianOS, and OS/2, but these are not officially supported.
I was guessing that although there is nothing even remotely new about a.csv file, our reviewer mentions this as an example of good style. You could write data out in any format you invent for each project, and with Python you might be particularly tempted to use some pickle (Python's object serialization) format. But using CSV whenever it makes any sense seems like it would give you the most long-term flexibility since so very many programs can read these files, even text editors. The format is never going away because it's too simple.
The separator you use is a non-issue since almost any program that reads these in can understand most typical separators. Excel might be a rare exception, I couldn't see an obvious way to specify that but I'm sure someone has a plugin that works. Tabs are okay too, and gnuplot would like those, but I'd still call it a CSV file just to be boring.
Of course when it comes to Christianity there are lots of ways of responding to what's in the Bible. I like to see ethics in the Bible as an evolution of thought throughout history. For the time they were written, the laws you quote were (probably) much more fair and compassionate than laws they superseded, or laws of the other cultures at that time. I don't know if that's specifically true of what you are quoting or not, but this is the case more often than not. And there is definitely sexism there, but again less so than in other cultures at the time. Many of these laws (such as the ten commandments) were completely revolutionary at the time, and I think Jesus' ideas continued in the same direction. They were (still are) shockingly revolutionary, defied "common" sense, and stretched people toward a better sense of morality.
But this thread is really about the shameful behavior of some church officials and whether or not to blame the church for hiding their actions. These coverups seem outrageous to me too, but maybe some of that is because I'm too disconnected from the actual events and people involved, and all i can see is the headlines and accusations. So for me it looks like a very black and white issue, and I have that comfortable, smug feeling of self-righteousness knowing that I'm not a pedophile and I am not hiding one in my house. This would make me happy if I didn't know that most of the time this feeling comes from my own misunderstanding of the real situation.
Probably this issue is really about reformation -- forgiveness and/or reformation of sin (in Christian terms) or the correctional system (in legal terms). Obviously church officials wanted do handle Geoghan in their own way, and the government in their way.
Okay I still side with the government on this because Geoghan would have had more contact with children when being moved from parish to parish, and I think there's a stronger imperative to protect others than to reform the criminal. But at least I can admit that the whole issue of reform is very complex and is divisive in politics and in religion. Other church officials probably believed his confessions were enough to solve the problem. But really, no one has any obvious solutions. Prisons do some good at detering crime and protecting the public, but are terrible when it comes to reformation. Eventually (most) prisoners are released, and it's hard to expect that they come out better off for the time served. In Christianity we have accounts of Jesus being forgiving, but at the same time demanding complete reformation as well. It's a tough act to follow, to say the least! I think we can help others in the same way, a little bit, but only to the extent that we ourselves are humble, selfless, kind, and such. Well, that's not my point, I'm rambling as usual.
My point is that it's way to easy to be a critical bystander and not want to really get into the issues and mind-set of the people involved. I'm both disconnected and completely uninformed about the many cases with Scientology, and I'm even more skeptical of any defense of theirs since the whole religion/cult seems to revolve around information control and the careful management of public appearance. But I've never actually studied scientology so once more I'm in no position to judge accurately. I don't intend to study it at the moment, so I'll have to leave all the bashing to others.
You're right about the compensation being too small to be worth considering in comparison with the possible inconvenience.
I would guess that a power company might really benefit from this kind of arrangement though, especially when trying to recover from a power failure. When they first try to bring power back on, everyone's A/C, refrigerators, and maybe even flourescent light starters draw a large amount of current. Many times this creates too much demand on the working (backup or alternative) power lines and they just have to wait until mechanical repairs are completed, many hours later. They tell you to turn off your AC and fridge after the power fails, but not many people actually do that. Everyone wants to be first, and so everyone loses.
I was reading this thread because I have a Gentoo box for stuff like web serving and mythtv, but would like the flexibility to let my kids play a Windows game on it now and then. These are slightly-obscure kids games that work poorly even under windows and are probably not worth the effort to get working under wine.
My question is whether or not I can run them on a virtual PC (xen, virtualbox, or kvm)? I get the impression that I can't since they inveriably use directX which isn't supported in a virtualized environment. But... am I wrong about that? I hope so.
I think Dunkirk has it exactly right here. While I can guess that his religious beliefs (which I don't presume to really know about) are at least not inconsistent with his post, you're trying to use religion as an excuse to dismiss the ideas automatically instead of addressing them on their own merit.
I too have been married for 15 years, and I think my relationship with my wife is strong because it's based on friendship and mutual support, not sex. We both enjoy sex but with kids and madly busy schedules, it would never be enough to keep us together. We both still reminisce about life before kids, when we could actually date now and then. Obviously if a better sex life was my main desire I'd drop my inconvenient commitments and run off with someone else. That has nothing to do with how good the sex was in the beginning of our relationship. If I were to have chosen to marry her for the sex, even if she were playgirl of the year then, I'd be dissatisfied by now.
We've been through a lot in the past year especially since I gave up what was pretty close to my dream job in order to support her professional needs and the needs of our kids. In other words, I think one needs to love their spouse so much that what benefits one's self is much less important than what benefits their family. She has done the same for me in the past.
I *know* I sound "preachy" here, but why? I'm not arguing that friendship trumps sex because of any kind of right relationship with God or, far worse, some fear of afterlife or hell. I guess the only reason this sounds like a religious argument is because we hear this same kind of message from the religious right wing. We then tend to accept or reject the idea based on other opinions not even remotely related to the relevant arguments. This same thing happens all the time in politics, and it bothers me.
I'm just trying to understand your (missing but implied?) argument. I'm not being obnoxious - I really want to know.
So you're suggesting that agents in a free market economy will be self-motivated to fund charities and public services like roads, law enforcement, basic research, etc? I'm missing a central piece of your argument because I thought the free market economy was based on the principle that one does the most good to the economy by maximizing their own profits, choosing the cheapest and best merchandise to buy, and so on.
I can see how this works fairly well for many markets such as food, furniture, beauticians, and I guess most manufacturing and personal service industries. But there are an increasing number of areas of work where the beneficiaries are not just the direct consumers. Who benefits from law enforcement or national defense? Who should pay for it? I don't see how a free market economy will self-regulate toward an optimal level of funding for this, or how it can determine who should be footing the bill?
One example, for me, of the failure of the free market to address these sectors is the steady loss of corporate funding for research. Bell labs is a good example, if you want a case study. It did a tremendous amount of good for the US and world economy, and produced a lot of important research in fields like computer science and even astrophysics. But ultimately this did not benefit its corporate sponsors enough, and it has been scaling down over the past decade or two. As market forces drive global competition to higher and higher levels, companies must either focus on their core business or fold.
That's my central argument. A kind of survival of the fittest principle is at work to help maximize the efficiency of companies today. This only tends to reduce corporate spending on programs that are of more public interest than private.
I am not, Not, NOT convinced that government spending and regulation is the ideal answer to this problem. But it works to some degree whereas relying on sacrificial corporate generosity does not work to any degree in the long term. I'm very interested in finding a market-driven solution that works and is self-regulating. But at the moment I have almost no idea what that would be like.
It's interesting to think of what industries are in or near this region where (I believe) a market economy looses its effectiveness. There are the obvious ones like public infrastructure, national defense, and basic research. Health care and insurance are less obvious. But free software is also in this category where market forces do not provide an optimal level of funding. I suspect that advertising is also an interesting market from this perspective, since while companies will want to spend money on advertising to increase public awareness of and desire for their products, they will not promote product information that is of value to consumers who want to make informed decisions. It's also not obvious to me how financial services are properly regulated by the market.
And in these sectors - software, advertising, and finance, it seems obvious that governmental sponsorship and regulation is also not efficient. I'm at a loss for a solution.
This seems like the right line of reasoning. Are beer companies taking carbon out of the ground and putting it in the air (in the form of CO2)?
But even if you just take CO2 production into account, this sounds like a perfect Fermi problem to me. I have seen a lot of posts claiming that CO2 release is minuscule compared to oil and coal burning, but nothing that hints at HOW minuscule.
We probably need to guess at total world alcohol production, world coal and oil burning, and guess how much CO2 is released manufacturing the alcohol. It's tempting to search the web, but the whole point of a Fermi problem is that you don't even have to make good guesses when all you're concerned about is powers of ten.
Alcohol production: maybe 1 beer per person per day,.3 L per bottle, 7 billion people: 10^12 L of beer per year, or 10^11 L of alcohol or 10^11 kg if by weight. Suppose the process produces as much CO2 by weight as alcohol. After all, what do I know? I feel like I've estimated things on the high side, especially bottles per person per day, but come up with 10^11 kg of CO2 per year. this is probably at least 10 times too high.
Coal and oil: this is much harder to estimate and much easier to look up. It's about 3 * 10^13 kg.
So CO2 release from beer production is not completely insignificant as I was guessing at first. I'd guess it's around 1/1000th as much as from fossil fuel consumption, give or take a power of ten.
So the first point now becomes more relevant. The alcohol in drinks is made from sugar, which comes from plants that take the carbon out of the atmosphere. So it's a closed cycle except for the electricity used for manufacturing and gasoline for distribution.
They also don't know anything, beyond reasonable guesses, about all the different ways a Higgs would decay and how often. These "branching ratios" would give lots of clues to physics at even deeper levels. Also there is the lifetime of the particle that can be measured by measuring the spread of the distribution of masses for many observed Higgs. It works by a kind of uncertainty principle -- the shorter the lifetime the larger the spread in mass.
I can only assume that the 50/50 chance quoted here is under the assumption that a "standard model" Higgs does exist, whose mass has been constrained by earlier experiments.
And no, we are definitely not approaching any kind of nadir in particle physics because there is still a logical incompatibility between particle physics (quantum field theory) and gravity. Really, neither theory can be true. I guess physicists in the pre-1905 era were less concerned about such things, but these days we have for some reason come to expect that a final theory can be proposed which would be internally consistent. There are also lots of nagging questions, or loose threads, about the standard model of particle physics. These issues are in some sense aesthetic, but at the same time they have inspired a very large number of new theories, most of which have yet to be disproved.
Probably the most interesting discovery at Fermilab or CERN would not be observing a Higgs particle, but instead seeing something else such as a new supersymmetric particle or something entirely unexpected. Both machines are probing an energy scale that is a kind of unification threshold for the electromagnetic and electroweak forces, and all kinds of speculative theories are ripe with predictions at this energy. The other unification thresholds are far, far beyond reach of today's experiments.
I think you're right about the scales here, but only according to general relativity (in 4 dimensions), which also predicts that black holes of any sort will not form at the LHC. If they could, their radius would be about 10^-50 m, which is small compared to atomic distances (10^-10 m) or even nuclear distances (10^-15 m), and requires far greater energy densities than the LHC can produce.
The hope is that by observing the rapid evaporation of black holes at the LHC, one could support a class of theories that predict "large" extra dimensions. I'm not sure if this is exclusive to string theory or not, but I think the basic idea is that if the universe is really 10-dimensional or so (which seems to be a general requirement of any string theory), the "extra" (not observed) dimensions must be somehow curled up in such a way that they have no affect on any observed phenomena. Large Extra Dimensions is the idea that some of these may not be curled up so tightly, and would give rise to new physics on very small scales. It would then be possible for microscopic black holes to be many orders of magnitude larger than one would expect otherwise, and possibly even within reach of whatever energy scale the latest accelerator is running at.
As I describe it here, this theory seems rather far-fetched. It is a weakly-motivated extension of string theory, and all string theory has going for it right now is a peculiar aesthetic. But I'm sure I'm not doing it enough justice. I think Large Extra Dimensions is an attractive theory because it helps resolve quite a few problems of consistency and scale (or "naturalness") in more standard theories. But why the scale of these extra dimensions should be just enough so that we could start to see effects of it at the LHC and not at even higher energies is particularly hard to justify.
So seeing black holes at the LHC seems extremely unlikely, but it's worth looking for them because the importance of such a discovery would be huge.
Then there's the problem of whether or not these black holes could somehow rapidly grow (not evaporate) and shrink the earth down to the size of a marble. Well, I'd think that if this were the case, we'd probably have seen "harmless" black holes at lower energies already, such as at RHIC. They looked, but found nothing. And then there's the cosmic ray argument that others have mentioned here. Cosmic ray collisions at energies at (and way beyond) the LHC energies have been occurring all the time, and at rates much higher than the LHC as well. I think there are lots of observations that prove our universe is not so delicate.
It sounds to me like the article referenced does not take into account any kind of far-fetchedness of a doomsday theory. Maybe we're living in a matrix-like virtual world and with just the wrong combination of words typed into my computer, or the wrong set of thoughts in my mind, I'd expose a new bug in The Simulator and crash it, destroying all life as we know it. There is a lot of uncertainty in this theory. Does that mean we must live in constant fear? If we must, there are better things to worry about. So for me this "Probing the Improbable" article fails to pass a sanity test.
This is based on a highly speculative theory with absolutely no data to back it up. Not that it isn't an interesting theory, but please realize how premature it is! There are literally hundreds of other competing theories out there, and I'm sure we haven't even begun to get really creative. It's unfortunate that experimental particle physics seems to be about 20 years behind theory. In part I blame the SSC cancellation for this. Theorists have had little more than aesthetics to go on lately.
I think the advice to set swap at 2x RAM is a very rough guide based on having a typical desktop setup that has been optimized with about as much RAM as it needs. So from that point of view if you have 8GB RAM then you probably also are running very demanding apps that use it, and some more swap might help during usage spikes.
The other issue is using RAM for disk caches. The more you store in cache, the better your performance (to a point). So if some rarely-used data sits in memory, it might be better to swap it out and use that much more memory for cache. Again this assumes you've chosen your RAM size for optimal price/performance.
I thought CFLs and LEDs gained good efficiencies because the visible spectrum is small compared to usable blackbody spectra. LEDs and CFLs have line emissions that are entirely (or mostly) in the visible spectrum. A full-spectrum light source might be efficient if you integrate over all frequencies, but I'm sure a large fraction of this emission will be in the UV or IR ranges (86% I believe). For our eyes, this is less efficient. See http://en.wikipedia.org/wiki/Luminous_efficacy
I tried out votebyissue.org, and found that I got somewhat helpful results by scoring candidates not on the issues but how well they stated them. Most candidates got negative scores because they used sweeping generalities, drastically partisan statements, or just a lot of BS. I would agree if the statement was clear, supported, and at least somewhat reasonable. Yes, a few did get positive scores by the end, but only because I was unreasonably forgiving. I don't really think that anyone with enough desire for power to bid successfully for US president is fit to lead. But some are obviously worse than others. I guess I'm suggesting that we could do worse than to rate presidential candidates by how much or little they (ab)use propaganda.
I very much appreciated the ability to see statements without knowing which candidate gave them. However my method is very sensitive to exactly how the candidates represented themselves. Many statements were provided by someone from the candidates' campaigns, but it seemed that about half the positions were quotations from speeches. These quotations usually got artificially negative scores because people speak less precisely than they write. And yes, I would have preferred the option to give rankings instead of thumbs up, down, or neutral. Seeing the results in a table would have been nice, too.
On the other hand, presidents can definitely influence policy issues by threat of veto and influencing judicial appointments. Bush quite successfully devistated basic science funding in FY2008 with the threat to veto any spending bill over a certain amount. Thanks to blunders in both the house and senate, and outrageous strategies of party-vs-party mutually-assured destruction, this promised veto catalyzed a tragic loss of funding for almost all science. If current appropriations don't get reversed this year somehow, basic science in the US will probably be set back about 10-20 years because of our inability to sustain funding of long-term projects and our unwillingness to be a responsible member of international collaborations.
If you RTFA, you'll see that this guy violated Paint.NET's current license, so putting a different license in there would solve absolutely nothing.
But the GPL has been "tested" in court, while Paint.NET's current license has, I assume, not been yet. Also there are organizations that will help you in court if it's a GPL violation. So in part it's a matter of practicality, not principle.
Also Paint.NET should consider exactly how they want legitimately derived works to happen. If the GPL prevents certain kinds of derived works that they might like to see others create, then it's not the right license on principle.
Hmm, currently they're using the MIT license, which is extremely permissive. I don't even see a prohibition against re-branding and re-crediting in the license. So it's not obvious to me that the current license is being violated. Perhaps it is and I'm just not seeing it because IANAL. Anyway, consider that the current license provides next to nothing in terms of protection, and that's what the authors chose. The GPL provides substantial protection against abuses, and if paint.net wants to whine, they should "sublicense" (which is explicitly permissible under the MIT license) first to demonstrate that they really don't want this stuff to happen. The MIT license looks to me like a big "kick me" sign.
Right, the second set of arguments were not "mathematical", but they were sequential. I don't think the submitter thought the argument was a mathematical one just because he had introduced "N". Instead he structured his arguments into a logical sequence. In some vague sense this is more mathematical since some part of mathematics deals with this form of logical argument. Think of the steps of algebraic manipulation, or lines of a geometrical proof. I think the work of Turing and others was also along these lines.
But I agree that the sequence was highly flawed in many ways. The submitter was struggling for ways to apply absolute principles to a situation that was more a matter of compromise -- trying to do more good than harm. He rejects all arguments based on responsibility. _Why_ is it right to ban smoking for people under 18? He doesn't see that the reasons for younger people to not smoke are substantially different than reasons for older people not to smoke. And these reasons are not necessarily true for every individual in every situation, but based on an aggregate effect, the smoking ban for minors may be doing some good. There is also the political environment to consider -- it would be very hard for politicians to pass a ban on smoking for adults as most politicians themselves smoke. But politicians aren't minors. The fact that smoking cannot be banned for adults does not mean that banning smoking for minors is wrong to do. Also, we allow adults to work in hazardous areas (with heavy machinery, construction sites, etc.) where children are not allowed to work. But sometimes adults get seriously injured or killed while working. Yet the same arguments used to ban children from hazardous work areas do not apply equally to adults. It just isn't practical (or necessary) to ban adults from this. How about military service? And so on...
I disagree with the logic leading to statements 1, 2, 4, 5, and 6. Statement 3 is a new axiom, and is valid only if you identify just one bad effect (such as lung cancer), and allow for uncertainties in N. But (3) is still problematic because it refers to averages, which encourages us to consider every individual to be average. This leads to lots more logical fallacies ("the flaw of averages").
One thing I think is lacking from most math courses is the use of math as a language. I believe that this can help to make the subject more relevant.
Usually math is taught as a collection of skills for algebraic manipulation. Most of my physics students are very good at doing the algebra (or calculus, etc.) once they have the right set of equations and know what to solve for. Once the problem is posed as a "math" problem, it's all easy for them. Well, most of the time.
While I'm happy with that by itself, students have a very hard time coming up with the equations to solve in the first place. The process of taking a (in my case) physical system and describing it in terms of mathematics is the major hurdle for most students. And that process -- describing ideas in terms of mathematics -- is exactly why math is practical, relevant, memorable, and fun. Some ideas that can be expressed in mathematics are proportionality, balance, rate, and approximation. These are purely abstract concepts until they are applied to specific situations, which makes the language kind of tricky to learn and also makes mathematics so universal. And I think this also explains why it's so rarely taught -- it's very unlike spoken languages that are designed to deal with specifics. The few students who are good at math are either naturally attuned to thinking about the math as ideas, or are amused by the puzzle-like nature of symbolic manipulation. The vast majority come to the conclusion that math is just hard. As the previous post points out, lots of things are "hard." But if you're motivated they seem worthwhile, which makes them seem easy. Algebraic manipulation is pitiful motivation by itself.
Abstract algebra (group theory) could also probably be taught at the high school level, simply because it greatly broadens the vocabulary. This subject makes it even easier to see how ideas are behind the symbols and operations.
I jokingly told some math professors that a good mathematics proficiency exam for incoming freshmen would contain just one question: Do you think you'll ever *actually use* all that math you learned in high school?
*Now* I know why I've been teaching my introductory physics class to solve "Fermi problems." It's just in case they get on a jury and have to reason through the order of magnitude of reasonable losses!:) Good lawyers play lots of mind games (bad math) when it comes to settlements.
IANAA but it seems quite possible that some explanation like rings or actively-growing clouds around a planet could explain this phenomena. For saturn-like rings to add much to the planet's apparent size though, they would have to be rotating at a steep angle with respect to the ecliptic plane. I can't imagine how a planet that close to its sun could get so skewed.
Hmm, I also wonder if scientists are taking into account other heating methods of the planet -- if it's hot enough it could be as large as it pleases. And if it's so big already and so near the sun, maybe tidal motion would heat the planet further? Well I can speculate like this all day but it's no good if I can't do any quantitative studies. And I can't.
Driver support in Linux is a kind of chicken-and-egg problem, of course. Until Linux gets much wider adoption (say 30% or more), hardware companies will keep dragging their feet with drivers. But many people are unwilling to try Linux ultimately because some piece of hardware is not supported (at all or not well enough). How can we get around this problem? Is there a solution?
I'm happily using Ubuntu on my old Dell Latitude, which actually got all the traditionally-tricky hardware stuff right first time, including wireless networking! Setting this up with WPA was even easier than in Windows! So when hardware just happens to be well-supported, the whole experience can be far superior to Windows.
I am also reminded of a time I upgraded an old computer that dual-booted between Windows 98 and Linux (Gentoo IIRC). I effectively put the old hard drive into a new computer. Linux booted up with no problems whatsoever. Zero work required, all the hardware was detected and configured. (I went from one NVIDIA video card to another, which helped that aspect.) Windows 98 was much less happy. I had to remove all hardware (including every aspect of the chipset) from the device manager at least three times, download dozens of drivers, and dig into online help forums to get it all working again. About four hours and twenty reboots later, success!
Finally, when people talk about how much easier Windows is than Linux they aren't taking into consideration how much they really know about Windows, and how much less they know about Linux. If I hear about how an airplane pilot has such a hard time running a submarine and how much easier airplanes are to use, I wouldn't take the report very seriously.
I haven't been moderating Slashdot comments recently, mostly because I don't feel like I can take the time to be fair about it. Maybe something along the lines of the article's suggestion would help. I don't think moderators should have their posts filtered by moderation. But then there are too many posts to read. So perhaps a moderator should be given a random sample of posts to moderate? Unfortunately you couldn't judge duplicates that way, and there would have to be some way of seeing parent posts. Moderators also could not judge whether or not a post is a duplicate. But I don't think that happens well enough as it is, so there's little harm in taking a broken feature and making it worse. Slashdot could also boost the frequency of appearance of newer posts, or better yet make posts with the least number of moderator views are the most likely to be selected. Someone in their profile could choose how many random posts to view when moderating.
I know, none of this resolves the problem that people tend to mod up posts that they agree with, as apposed to posts that are well-motivated but disagreeable. But that's deeply-rooted human nature -- people love to hear about how right they are -- and I can't imagine how any automated system could compensate. At the least Slashdot could check for internal plagarism without too much trouble, given enough CPU cycles.
I read the impeachment summary too, and liked it. But I don't think this will go anywhere. I'm jaded. I remember when this news came out, slashdotted in the NY Times I think, well before the 2004 presidential election. I remember thinking that this was an impeachable offense, and telling other conservative friends and family members about it. They did not believe me, and/or they didn't think it was really that bad. At least it wasn't nearly as bad in their opinions as the accusations in "Unfit for Command." I did not vote for Bush/Cheney, of course. But enough people did vote for them, and here we are. The second best reason I had for my vote was the shutdown the national nuclear security administration advisory committee back in 2003 after they proved (in public) that "bunker buster" nukes could never hope to destroy a deeply buried target, and could never contain the explosion and fallout. They also proved that the proposed neutron bomb for neutralizing bio-weapons was a far better delivery mechanism than anything else. To respond to these findings by disbanding the advisory committee was not good.
With Bush's reelection, I lost my last shred of hope for politics.
Unfortunately, the US also has "egg on their face" from other goings-on in particle physics. Another fairly recent disaster was the cancellation of BTeV. This was most unfortunate because European collaborators were completely disenfranchised. By not having a system in place that can effectively fund a multi-year research project, we've lost valuable collaborators and lost international credibility. In addition to this, we've lost enormous amounts of funding for particle physics over the past decade, and as of now there are no major new experiments being built in the US, and everything that's running will pretty much shut down by 2008 (Fermilab, SLAC, Brookhaven, CESR/CLEO). All Fermilab has going for it after 2008 is that they can build magnets, and now with these issues maybe even that is suspect. As particle physics tends to thrive only on relatively large experiments that take well over a decade to go from proposal to construction and finally operation, it's hard to imagine that basic science in the US will even be relevant any more to the worldwide community for at least the next few decades, if ever again. What's just as frustrating as this was the complete lack of media coverage as the US accomplished its "exit strategy" in particle physics, beginning in about 1993 and ending just about now.
Slashdot Mirror
While I basically agree with you, in that programming is a task best done well or not at all, it's a common mistake to think that everyone who programs should be a career programmer. It must be possible to teach the necessary programming skills (and they are necessary) to college students in engineering, science, and a few other fields. It's a little bit like mathematics except that many students don't start learning it early enough. If one doesn't start learning programming before college, then it's possibly too late to really learn enough in college. I think these days really successful students primarily learn it outside of formal classes.
My take on the topic of Fortran is that it is not important to learn in school. I've written a lot of Fortran, but with a strong background in C it was not hard to learn when it was necessary for me to. Even in experimental physics, Fortran is used increasingly rarely so most students would not have a need to use it.
I think that learning one programming language in school is not nearly enough. Students should learn basic procedural programming concepts, something about data structures, object oriented design, and good practices in some scripting language to get comfortable with hashes and regular expressions. Python is great for a lot of this, but it's probably better to establish more low-level concepts first, including pointers. Maybe learning C and then Python could be a reasonable, minimalistic approach? C++ could actually be avoided since the syntax and rules are so complex. (I'm not talking about comp-sci majors who should certainly learn C++.) Perl is awesome of course, but again the syntax is complex and in the context of a college course that just takes time away from learning the essential concepts. Java is possible but doesn't really get either low-level or high-level enough for my tastes.
Whatever a student learns, they should have enough breadth, conceptual knowledge, and practice so that whatever language is required for a given project would not be too hard to learn on the job.
I think you're right about the acoustics problem being non-trivial. There are many factors to consider in the design of instruments. I'd guess that the first problem to solve is how one can reproduce the notes called for in Bach's score. The shape of the mouthpiece, bore, and bell will affect the pitches that it resonates at. Real brass instruments play notes in a kind of harmonic sequence, and notes get closer together as you go up the scale. Sounds like this instrument is a little like a french horn in that it's about the same length (about the same as a trombone too) but is played high (maybe 3 octaves or more?) above the fundamental. But the exact shape of the instrument is what dictates the frequencies of these modes of vibration. Probably even the stiffness of the brass makes a small difference. And beyond the notes it can make, each resonant mode will have a kind of width, or Q value, that relates to how easy it is to bend the pitch to something in a 12-tone scale. A related problem is working out the timbre or general sound of the instrument, also a function of the shape. I don't know how the designers would have any reference for that.
I don't know how they went about modeling the instrument. I guess the basic approach would be to consider a geometry and then sweep through frequencies, looking for resonances. But how would you simulate the buzzing of lips in a mouthpiece? If they just put a microphone up to a mouthpiece of some shape and had a musician play different pitches, that still wouldn't account for the feedback or back-pressure of the rest of the instrument.
While I know enough physics and acoustics to be arrogant, I still would hardly know where to start on such a project. I'm really curious to learn about the details of what they did.
From http://www.libsdl.org/ :
SDL supports Linux, Windows, Windows CE, BeOS, MacOS, Mac OS X, FreeBSD, NetBSD, OpenBSD, BSD/OS, Solaris, IRIX, and QNX. The code contains support for AmigaOS, Dreamcast, Atari, AIX, OSF/Tru64, RISC OS, SymbianOS, and OS/2, but these are not officially supported.
You just might have overreacted slightly.
I was guessing that although there is nothing even remotely new about a .csv file, our reviewer mentions this as an example of good style. You could write data out in any format you invent for each project, and with Python you might be particularly tempted to use some pickle (Python's object serialization) format. But using CSV whenever it makes any sense seems like it would give you the most long-term flexibility since so very many programs can read these files, even text editors. The format is never going away because it's too simple.
The separator you use is a non-issue since almost any program that reads these in can understand most typical separators. Excel might be a rare exception, I couldn't see an obvious way to specify that but I'm sure someone has a plugin that works. Tabs are okay too, and gnuplot would like those, but I'd still call it a CSV file just to be boring.
Of course when it comes to Christianity there are lots of ways of responding to what's in the Bible. I like to see ethics in the Bible as an evolution of thought throughout history. For the time they were written, the laws you quote were (probably) much more fair and compassionate than laws they superseded, or laws of the other cultures at that time. I don't know if that's specifically true of what you are quoting or not, but this is the case more often than not. And there is definitely sexism there, but again less so than in other cultures at the time. Many of these laws (such as the ten commandments) were completely revolutionary at the time, and I think Jesus' ideas continued in the same direction. They were (still are) shockingly revolutionary, defied "common" sense, and stretched people toward a better sense of morality.
But this thread is really about the shameful behavior of some church officials and whether or not to blame the church for hiding their actions. These coverups seem outrageous to me too, but maybe some of that is because I'm too disconnected from the actual events and people involved, and all i can see is the headlines and accusations. So for me it looks like a very black and white issue, and I have that comfortable, smug feeling of self-righteousness knowing that I'm not a pedophile and I am not hiding one in my house. This would make me happy if I didn't know that most of the time this feeling comes from my own misunderstanding of the real situation.
Probably this issue is really about reformation -- forgiveness and/or reformation of sin (in Christian terms) or the correctional system (in legal terms). Obviously church officials wanted do handle Geoghan in their own way, and the government in their way.
Okay I still side with the government on this because Geoghan would have had more contact with children when being moved from parish to parish, and I think there's a stronger imperative to protect others than to reform the criminal. But at least I can admit that the whole issue of reform is very complex and is divisive in politics and in religion. Other church officials probably believed his confessions were enough to solve the problem. But really, no one has any obvious solutions. Prisons do some good at detering crime and protecting the public, but are terrible when it comes to reformation. Eventually (most) prisoners are released, and it's hard to expect that they come out better off for the time served. In Christianity we have accounts of Jesus being forgiving, but at the same time demanding complete reformation as well. It's a tough act to follow, to say the least! I think we can help others in the same way, a little bit, but only to the extent that we ourselves are humble, selfless, kind, and such. Well, that's not my point, I'm rambling as usual.
My point is that it's way to easy to be a critical bystander and not want to really get into the issues and mind-set of the people involved. I'm both disconnected and completely uninformed about the many cases with Scientology, and I'm even more skeptical of any defense of theirs since the whole religion/cult seems to revolve around information control and the careful management of public appearance. But I've never actually studied scientology so once more I'm in no position to judge accurately. I don't intend to study it at the moment, so I'll have to leave all the bashing to others.
You're right about the compensation being too small to be worth considering in comparison with the possible inconvenience.
I would guess that a power company might really benefit from this kind of arrangement though, especially when trying to recover from a power failure. When they first try to bring power back on, everyone's A/C, refrigerators, and maybe even flourescent light starters draw a large amount of current. Many times this creates too much demand on the working (backup or alternative) power lines and they just have to wait until mechanical repairs are completed, many hours later. They tell you to turn off your AC and fridge after the power fails, but not many people actually do that. Everyone wants to be first, and so everyone loses.
I was reading this thread because I have a Gentoo box for stuff like web serving and mythtv, but would like the flexibility to let my kids play a Windows game on it now and then. These are slightly-obscure kids games that work poorly even under windows and are probably not worth the effort to get working under wine.
My question is whether or not I can run them on a virtual PC (xen, virtualbox, or kvm)? I get the impression that I can't since they inveriably use directX which isn't supported in a virtualized environment. But ... am I wrong about that? I hope so.
This is far off topic but I'll bite.
I think Dunkirk has it exactly right here. While I can guess that his religious beliefs (which I don't presume to really know about) are at least not inconsistent with his post, you're trying to use religion as an excuse to dismiss the ideas automatically instead of addressing them on their own merit.
I too have been married for 15 years, and I think my relationship with my wife is strong because it's based on friendship and mutual support, not sex. We both enjoy sex but with kids and madly busy schedules, it would never be enough to keep us together. We both still reminisce about life before kids, when we could actually date now and then. Obviously if a better sex life was my main desire I'd drop my inconvenient commitments and run off with someone else. That has nothing to do with how good the sex was in the beginning of our relationship. If I were to have chosen to marry her for the sex, even if she were playgirl of the year then, I'd be dissatisfied by now.
We've been through a lot in the past year especially since I gave up what was pretty close to my dream job in order to support her professional needs and the needs of our kids. In other words, I think one needs to love their spouse so much that what benefits one's self is much less important than what benefits their family. She has done the same for me in the past.
I *know* I sound "preachy" here, but why? I'm not arguing that friendship trumps sex because of any kind of right relationship with God or, far worse, some fear of afterlife or hell. I guess the only reason this sounds like a religious argument is because we hear this same kind of message from the religious right wing. We then tend to accept or reject the idea based on other opinions not even remotely related to the relevant arguments. This same thing happens all the time in politics, and it bothers me.
I'm just trying to understand your (missing but implied?) argument. I'm not being obnoxious - I really want to know.
So you're suggesting that agents in a free market economy will be self-motivated to fund charities and public services like roads, law enforcement, basic research, etc? I'm missing a central piece of your argument because I thought the free market economy was based on the principle that one does the most good to the economy by maximizing their own profits, choosing the cheapest and best merchandise to buy, and so on.
I can see how this works fairly well for many markets such as food, furniture, beauticians, and I guess most manufacturing and personal service industries. But there are an increasing number of areas of work where the beneficiaries are not just the direct consumers. Who benefits from law enforcement or national defense? Who should pay for it? I don't see how a free market economy will self-regulate toward an optimal level of funding for this, or how it can determine who should be footing the bill?
One example, for me, of the failure of the free market to address these sectors is the steady loss of corporate funding for research. Bell labs is a good example, if you want a case study. It did a tremendous amount of good for the US and world economy, and produced a lot of important research in fields like computer science and even astrophysics. But ultimately this did not benefit its corporate sponsors enough, and it has been scaling down over the past decade or two. As market forces drive global competition to higher and higher levels, companies must either focus on their core business or fold.
That's my central argument. A kind of survival of the fittest principle is at work to help maximize the efficiency of companies today. This only tends to reduce corporate spending on programs that are of more public interest than private.
I am not, Not, NOT convinced that government spending and regulation is the ideal answer to this problem. But it works to some degree whereas relying on sacrificial corporate generosity does not work to any degree in the long term. I'm very interested in finding a market-driven solution that works and is self-regulating. But at the moment I have almost no idea what that would be like.
It's interesting to think of what industries are in or near this region where (I believe) a market economy looses its effectiveness. There are the obvious ones like public infrastructure, national defense, and basic research. Health care and insurance are less obvious. But free software is also in this category where market forces do not provide an optimal level of funding. I suspect that advertising is also an interesting market from this perspective, since while companies will want to spend money on advertising to increase public awareness of and desire for their products, they will not promote product information that is of value to consumers who want to make informed decisions. It's also not obvious to me how financial services are properly regulated by the market.
And in these sectors - software, advertising, and finance, it seems obvious that governmental sponsorship and regulation is also not efficient. I'm at a loss for a solution.
This seems like the right line of reasoning. Are beer companies taking carbon out of the ground and putting it in the air (in the form of CO2)?
But even if you just take CO2 production into account, this sounds like a perfect Fermi problem to me. I have seen a lot of posts claiming that CO2 release is minuscule compared to oil and coal burning, but nothing that hints at HOW minuscule.
We probably need to guess at total world alcohol production, world coal and oil burning, and guess how much CO2 is released manufacturing the alcohol. It's tempting to search the web, but the whole point of a Fermi problem is that you don't even have to make good guesses when all you're concerned about is powers of ten.
Alcohol production: maybe 1 beer per person per day, .3 L per bottle, 7 billion people: 10^12 L of beer per year, or 10^11 L of alcohol or 10^11 kg if by weight. Suppose the process produces as much CO2 by weight as alcohol. After all, what do I know? I feel like I've estimated things on the high side, especially bottles per person per day, but come up with 10^11 kg of CO2 per year. this is probably at least 10 times too high.
Coal and oil: this is much harder to estimate and much easier to look up. It's about 3 * 10^13 kg.
So CO2 release from beer production is not completely insignificant as I was guessing at first. I'd guess it's around 1/1000th as much as from fossil fuel consumption, give or take a power of ten.
So the first point now becomes more relevant. The alcohol in drinks is made from sugar, which comes from plants that take the carbon out of the atmosphere. So it's a closed cycle except for the electricity used for manufacturing and gasoline for distribution.
They also don't know anything, beyond reasonable guesses, about all the different ways a Higgs would decay and how often. These "branching ratios" would give lots of clues to physics at even deeper levels. Also there is the lifetime of the particle that can be measured by measuring the spread of the distribution of masses for many observed Higgs. It works by a kind of uncertainty principle -- the shorter the lifetime the larger the spread in mass.
I can only assume that the 50/50 chance quoted here is under the assumption that a "standard model" Higgs does exist, whose mass has been constrained by earlier experiments.
And no, we are definitely not approaching any kind of nadir in particle physics because there is still a logical incompatibility between particle physics (quantum field theory) and gravity. Really, neither theory can be true. I guess physicists in the pre-1905 era were less concerned about such things, but these days we have for some reason come to expect that a final theory can be proposed which would be internally consistent. There are also lots of nagging questions, or loose threads, about the standard model of particle physics. These issues are in some sense aesthetic, but at the same time they have inspired a very large number of new theories, most of which have yet to be disproved.
Probably the most interesting discovery at Fermilab or CERN would not be observing a Higgs particle, but instead seeing something else such as a new supersymmetric particle or something entirely unexpected. Both machines are probing an energy scale that is a kind of unification threshold for the electromagnetic and electroweak forces, and all kinds of speculative theories are ripe with predictions at this energy. The other unification thresholds are far, far beyond reach of today's experiments.
I think you're right about the scales here, but only according to general relativity (in 4 dimensions), which also predicts that black holes of any sort will not form at the LHC. If they could, their radius would be about 10^-50 m, which is small compared to atomic distances (10^-10 m) or even nuclear distances (10^-15 m), and requires far greater energy densities than the LHC can produce.
The hope is that by observing the rapid evaporation of black holes at the LHC, one could support a class of theories that predict "large" extra dimensions. I'm not sure if this is exclusive to string theory or not, but I think the basic idea is that if the universe is really 10-dimensional or so (which seems to be a general requirement of any string theory), the "extra" (not observed) dimensions must be somehow curled up in such a way that they have no affect on any observed phenomena. Large Extra Dimensions is the idea that some of these may not be curled up so tightly, and would give rise to new physics on very small scales. It would then be possible for microscopic black holes to be many orders of magnitude larger than one would expect otherwise, and possibly even within reach of whatever energy scale the latest accelerator is running at.
As I describe it here, this theory seems rather far-fetched. It is a weakly-motivated extension of string theory, and all string theory has going for it right now is a peculiar aesthetic. But I'm sure I'm not doing it enough justice. I think Large Extra Dimensions is an attractive theory because it helps resolve quite a few problems of consistency and scale (or "naturalness") in more standard theories. But why the scale of these extra dimensions should be just enough so that we could start to see effects of it at the LHC and not at even higher energies is particularly hard to justify.
So seeing black holes at the LHC seems extremely unlikely, but it's worth looking for them because the importance of such a discovery would be huge.
Then there's the problem of whether or not these black holes could somehow rapidly grow (not evaporate) and shrink the earth down to the size of a marble. Well, I'd think that if this were the case, we'd probably have seen "harmless" black holes at lower energies already, such as at RHIC. They looked, but found nothing. And then there's the cosmic ray argument that others have mentioned here. Cosmic ray collisions at energies at (and way beyond) the LHC energies have been occurring all the time, and at rates much higher than the LHC as well. I think there are lots of observations that prove our universe is not so delicate.
It sounds to me like the article referenced does not take into account any kind of far-fetchedness of a doomsday theory. Maybe we're living in a matrix-like virtual world and with just the wrong combination of words typed into my computer, or the wrong set of thoughts in my mind, I'd expose a new bug in The Simulator and crash it, destroying all life as we know it. There is a lot of uncertainty in this theory. Does that mean we must live in constant fear? If we must, there are better things to worry about. So for me this "Probing the Improbable" article fails to pass a sanity test.
This is based on a highly speculative theory with absolutely no data to back it up. Not that it isn't an interesting theory, but please realize how premature it is! There are literally hundreds of other competing theories out there, and I'm sure we haven't even begun to get really creative. It's unfortunate that experimental particle physics seems to be about 20 years behind theory. In part I blame the SSC cancellation for this. Theorists have had little more than aesthetics to go on lately.
I think the advice to set swap at 2x RAM is a very rough guide based on having a typical desktop setup that has been optimized with about as much RAM as it needs. So from that point of view if you have 8GB RAM then you probably also are running very demanding apps that use it, and some more swap might help during usage spikes.
The other issue is using RAM for disk caches. The more you store in cache, the better your performance (to a point). So if some rarely-used data sits in memory, it might be better to swap it out and use that much more memory for cache. Again this assumes you've chosen your RAM size for optimal price/performance.
I thought CFLs and LEDs gained good efficiencies because the visible spectrum is small compared to usable blackbody spectra. LEDs and CFLs have line emissions that are entirely (or mostly) in the visible spectrum. A full-spectrum light source might be efficient if you integrate over all frequencies, but I'm sure a large fraction of this emission will be in the UV or IR ranges (86% I believe). For our eyes, this is less efficient. See http://en.wikipedia.org/wiki/Luminous_efficacy
Wow, I'd mod parent up to "6, True" if I could!
I tried out votebyissue.org, and found that I got somewhat helpful results by scoring candidates not on the issues but how well they stated them. Most candidates got negative scores because they used sweeping generalities, drastically partisan statements, or just a lot of BS. I would agree if the statement was clear, supported, and at least somewhat reasonable. Yes, a few did get positive scores by the end, but only because I was unreasonably forgiving. I don't really think that anyone with enough desire for power to bid successfully for US president is fit to lead. But some are obviously worse than others. I guess I'm suggesting that we could do worse than to rate presidential candidates by how much or little they (ab)use propaganda.
I very much appreciated the ability to see statements without knowing which candidate gave them. However my method is very sensitive to exactly how the candidates represented themselves. Many statements were provided by someone from the candidates' campaigns, but it seemed that about half the positions were quotations from speeches. These quotations usually got artificially negative scores because people speak less precisely than they write. And yes, I would have preferred the option to give rankings instead of thumbs up, down, or neutral. Seeing the results in a table would have been nice, too.
On the other hand, presidents can definitely influence policy issues by threat of veto and influencing judicial appointments. Bush quite successfully devistated basic science funding in FY2008 with the threat to veto any spending bill over a certain amount. Thanks to blunders in both the house and senate, and outrageous strategies of party-vs-party mutually-assured destruction, this promised veto catalyzed a tragic loss of funding for almost all science. If current appropriations don't get reversed this year somehow, basic science in the US will probably be set back about 10-20 years because of our inability to sustain funding of long-term projects and our unwillingness to be a responsible member of international collaborations.
Okay I'm done now.
But the GPL has been "tested" in court, while Paint.NET's current license has, I assume, not been yet. Also there are organizations that will help you in court if it's a GPL violation. So in part it's a matter of practicality, not principle.
Also Paint.NET should consider exactly how they want legitimately derived works to happen. If the GPL prevents certain kinds of derived works that they might like to see others create, then it's not the right license on principle.
Hmm, currently they're using the MIT license, which is extremely permissive. I don't even see a prohibition against re-branding and re-crediting in the license. So it's not obvious to me that the current license is being violated. Perhaps it is and I'm just not seeing it because IANAL. Anyway, consider that the current license provides next to nothing in terms of protection, and that's what the authors chose. The GPL provides substantial protection against abuses, and if paint.net wants to whine, they should "sublicense" (which is explicitly permissible under the MIT license) first to demonstrate that they really don't want this stuff to happen. The MIT license looks to me like a big "kick me" sign.
Right, the second set of arguments were not "mathematical", but they were sequential. I don't think the submitter thought the argument was a mathematical one just because he had introduced "N". Instead he structured his arguments into a logical sequence. In some vague sense this is more mathematical since some part of mathematics deals with this form of logical argument. Think of the steps of algebraic manipulation, or lines of a geometrical proof. I think the work of Turing and others was also along these lines.
But I agree that the sequence was highly flawed in many ways. The submitter was struggling for ways to apply absolute principles to a situation that was more a matter of compromise -- trying to do more good than harm. He rejects all arguments based on responsibility. _Why_ is it right to ban smoking for people under 18? He doesn't see that the reasons for younger people to not smoke are substantially different than reasons for older people not to smoke. And these reasons are not necessarily true for every individual in every situation, but based on an aggregate effect, the smoking ban for minors may be doing some good. There is also the political environment to consider -- it would be very hard for politicians to pass a ban on smoking for adults as most politicians themselves smoke. But politicians aren't minors. The fact that smoking cannot be banned for adults does not mean that banning smoking for minors is wrong to do. Also, we allow adults to work in hazardous areas (with heavy machinery, construction sites, etc.) where children are not allowed to work. But sometimes adults get seriously injured or killed while working. Yet the same arguments used to ban children from hazardous work areas do not apply equally to adults. It just isn't practical (or necessary) to ban adults from this. How about military service? And so on...
I disagree with the logic leading to statements 1, 2, 4, 5, and 6. Statement 3 is a new axiom, and is valid only if you identify just one bad effect (such as lung cancer), and allow for uncertainties in N. But (3) is still problematic because it refers to averages, which encourages us to consider every individual to be average. This leads to lots more logical fallacies ("the flaw of averages").
One thing I think is lacking from most math courses is the use of math as a language. I believe that this can help to make the subject more relevant.
Usually math is taught as a collection of skills for algebraic manipulation. Most of my physics students are very good at doing the algebra (or calculus, etc.) once they have the right set of equations and know what to solve for. Once the problem is posed as a "math" problem, it's all easy for them. Well, most of the time.
While I'm happy with that by itself, students have a very hard time coming up with the equations to solve in the first place. The process of taking a (in my case) physical system and describing it in terms of mathematics is the major hurdle for most students. And that process -- describing ideas in terms of mathematics -- is exactly why math is practical, relevant, memorable, and fun. Some ideas that can be expressed in mathematics are proportionality, balance, rate, and approximation. These are purely abstract concepts until they are applied to specific situations, which makes the language kind of tricky to learn and also makes mathematics so universal. And I think this also explains why it's so rarely taught -- it's very unlike spoken languages that are designed to deal with specifics. The few students who are good at math are either naturally attuned to thinking about the math as ideas, or are amused by the puzzle-like nature of symbolic manipulation. The vast majority come to the conclusion that math is just hard. As the previous post points out, lots of things are "hard." But if you're motivated they seem worthwhile, which makes them seem easy. Algebraic manipulation is pitiful motivation by itself.
Abstract algebra (group theory) could also probably be taught at the high school level, simply because it greatly broadens the vocabulary. This subject makes it even easier to see how ideas are behind the symbols and operations.
I jokingly told some math professors that a good mathematics proficiency exam for incoming freshmen would contain just one question: Do you think you'll ever *actually use* all that math you learned in high school?
*Now* I know why I've been teaching my introductory physics class to solve "Fermi problems." It's just in case they get on a jury and have to reason through the order of magnitude of reasonable losses! :) Good lawyers play lots of mind games (bad math) when it comes to settlements.
IANAA but it seems quite possible that some explanation like rings or actively-growing clouds around a planet could explain this phenomena. For saturn-like rings to add much to the planet's apparent size though, they would have to be rotating at a steep angle with respect to the ecliptic plane. I can't imagine how a planet that close to its sun could get so skewed.
Hmm, I also wonder if scientists are taking into account other heating methods of the planet -- if it's hot enough it could be as large as it pleases. And if it's so big already and so near the sun, maybe tidal motion would heat the planet further? Well I can speculate like this all day but it's no good if I can't do any quantitative studies. And I can't.
Driver support in Linux is a kind of chicken-and-egg problem, of course. Until Linux gets much wider adoption (say 30% or more), hardware companies will keep dragging their feet with drivers. But many people are unwilling to try Linux ultimately because some piece of hardware is not supported (at all or not well enough). How can we get around this problem? Is there a solution?
I'm happily using Ubuntu on my old Dell Latitude, which actually got all the traditionally-tricky hardware stuff right first time, including wireless networking! Setting this up with WPA was even easier than in Windows! So when hardware just happens to be well-supported, the whole experience can be far superior to Windows.
I am also reminded of a time I upgraded an old computer that dual-booted between Windows 98 and Linux (Gentoo IIRC). I effectively put the old hard drive into a new computer. Linux booted up with no problems whatsoever. Zero work required, all the hardware was detected and configured. (I went from one NVIDIA video card to another, which helped that aspect.) Windows 98 was much less happy. I had to remove all hardware (including every aspect of the chipset) from the device manager at least three times, download dozens of drivers, and dig into online help forums to get it all working again. About four hours and twenty reboots later, success!
Finally, when people talk about how much easier Windows is than Linux they aren't taking into consideration how much they really know about Windows, and how much less they know about Linux. If I hear about how an airplane pilot has such a hard time running a submarine and how much easier airplanes are to use, I wouldn't take the report very seriously.
I haven't been moderating Slashdot comments recently, mostly because I don't feel like I can take the time to be fair about it. Maybe something along the lines of the article's suggestion would help. I don't think moderators should have their posts filtered by moderation. But then there are too many posts to read. So perhaps a moderator should be given a random sample of posts to moderate? Unfortunately you couldn't judge duplicates that way, and there would have to be some way of seeing parent posts. Moderators also could not judge whether or not a post is a duplicate. But I don't think that happens well enough as it is, so there's little harm in taking a broken feature and making it worse. Slashdot could also boost the frequency of appearance of newer posts, or better yet make posts with the least number of moderator views are the most likely to be selected. Someone in their profile could choose how many random posts to view when moderating.
I know, none of this resolves the problem that people tend to mod up posts that they agree with, as apposed to posts that are well-motivated but disagreeable. But that's deeply-rooted human nature -- people love to hear about how right they are -- and I can't imagine how any automated system could compensate. At the least Slashdot could check for internal plagarism without too much trouble, given enough CPU cycles.
I read the impeachment summary too, and liked it. But I don't think this will go anywhere. I'm jaded. I remember when this news came out, slashdotted in the NY Times I think, well before the 2004 presidential election. I remember thinking that this was an impeachable offense, and telling other conservative friends and family members about it. They did not believe me, and/or they didn't think it was really that bad. At least it wasn't nearly as bad in their opinions as the accusations in "Unfit for Command." I did not vote for Bush/Cheney, of course. But enough people did vote for them, and here we are. The second best reason I had for my vote was the shutdown the national nuclear security administration advisory committee back in 2003 after they proved (in public) that "bunker buster" nukes could never hope to destroy a deeply buried target, and could never contain the explosion and fallout. They also proved that the proposed neutron bomb for neutralizing bio-weapons was a far better delivery mechanism than anything else. To respond to these findings by disbanding the advisory committee was not good.
With Bush's reelection, I lost my last shred of hope for politics.
Unfortunately, the US also has "egg on their face" from other goings-on in particle physics. Another fairly recent disaster was the cancellation of BTeV. This was most unfortunate because European collaborators were completely disenfranchised. By not having a system in place that can effectively fund a multi-year research project, we've lost valuable collaborators and lost international credibility. In addition to this, we've lost enormous amounts of funding for particle physics over the past decade, and as of now there are no major new experiments being built in the US, and everything that's running will pretty much shut down by 2008 (Fermilab, SLAC, Brookhaven, CESR/CLEO). All Fermilab has going for it after 2008 is that they can build magnets, and now with these issues maybe even that is suspect. As particle physics tends to thrive only on relatively large experiments that take well over a decade to go from proposal to construction and finally operation, it's hard to imagine that basic science in the US will even be relevant any more to the worldwide community for at least the next few decades, if ever again. What's just as frustrating as this was the complete lack of media coverage as the US accomplished its "exit strategy" in particle physics, beginning in about 1993 and ending just about now.