I agree with most of this, and certainly there is a lot to be learned from quick and dirty analyses, simple visualizations, etc. But one of the things that really concerns me about promoting the use of Excel for analysis (as I would imagine this book does) is that it lures researchers away from tools that are more appropriate for the job, but are more difficult to learn and use. Even though it will take people more time to reach the same level of fluency with R that they would have with Excel, I think it's time well spent. So while I appreciate the uses of quick and dirty analyses as much as the next guy, I'd still rather see them done in R than in Excel. Incidentally, I know that R is a little impenetrable to some users due to the lack of a comprehensive standard graphical interface. I only use it as an example because it's what I use, and I know it has the respect of serious stats jocks. Also, I think that anyone who's smart enough to know what kind of analysis they need should be able to learn R. Of course, for many of the more involved problems in many fields, you need software designed specifically for the job anyway, and neither R nor Excel nor any other general-purpose package will do.
On a slight tangent, one of the reasons I like openoffice (I ditched Word years ago, and the new release candidate is really pretty good) is that I try to avoid using it for things that neither it nor the Microsoft alternatives are good for. Charting is a great example, so I'm with you on that. I'm abysmal at charting in R, but it does tend to have what I need.
Care to expand on why you think you can't do 'meaningful scientific data analysis in Excel?' Are you one of these people who 'reviews' books without actually reading them?
Someone else has already posted a link to a page that nicely summarizes many (not all) of the problems with using excel for science. But there is virtually no statistical technique which isn't already better implemented in R (free) and many other statistical packages. Real stats packages provide implementations of a given technique that are at least as reliable, provide more control, more options, more diagnostics, and often more guidance. The built-in stuff in Excel is so oversimplified that I think if you're really forced to use it for serious statistics, you'd have to re-implement things using basic arithmetic operations. It's graphing capabilities, last time I checked, lacked the majority of even the most basic kinds of statistical/scientific graphs/plots. Sure, you can do this or that in Excel, and if you're willing to put in enough work you can often get what you really need out of it. But it's rarely if ever the best tool for the job of scientific data analysis.
I don't review books about reading them, and this is no exception. But I do have an informed opinion about the premise of this book (and to a lesser extent about the level of insight of a reviewer who seems, to put it mildly, easily awed). The premise that Excel is good to use for scientific data analysis is pretty deeply misguided. I'd be happy to be convinced otherwise if I were really wrong, but I can only set aside so much time for listening to arguments from nutcases (just in case one of them may have a point). I'm sure if I actually read this book, I'd learn about various useful things Excel can do that would surprise and impress me. But I already have all the information I need to form a reliable opinion on this question, and I value my time too much to read books about space aliens living among us or about doing analysis in Excel.
Truth to tell, I use openoffice calc (more or less an Excel clone) quite a bit for research-related things. But I'm careful with it, and don't rely on it for much more than moving numbers around.
There's something hard to reconcile about the reviewer's obvious awe and the fact that the book was written by someone who thinks doing meaningful scientific data analysis in Excel is a good idea.
Companies can suck and still produce products you'd like to use. I don't particularly like Apple (or Microsoft), but I do have reasons to want to run OSX (and Windows). One of the reasons I don't like Apple is because they make it difficult for me to support their platform, specifically by making it difficult to run on third-party or (more important to me) virtual hardware. I understand why, but understanding doesn't make me like them. There are lots of other reasons to dislike Apple, of course (e.g., their promotion of DRM-encrusted music), and very few of them are inconsistent with wanting to run OSX on the hardware of your choice. So for me, Apple does suck, and I'd still be pretty happy if I could run their OS more conveniently.
I just don't understand it when people say DVD is "good enough". You can see the compression artifacts! (and that's on a low resolution display)
DVD is absolutely good enough for me. All else being equal, of course I'd like better resolution. Sometimes I do find DVD artifacts bothersome, and I certainly resent it when they're poorly encoded. Of course, Blu-ray discs can be poorly encoded too, but it's liable to be less of a problem. Still, with my eyesight barely at 20/20 anymore, and my aging 32 inch LCD subtending no more than about 19 degrees of visual angle (that's at about 8 feet), the difference in resolution makes a lot less difference to me than to someone half my age watching a 50 inch set at the same distance. On top of which, when I watch movies, I tend to get caught up in things that are somewhat peripheral to the quality of the medium -- plot, acting, cinematography, etc. I've often watched movies with scratched film projected by crappy projectors. I often listen to audio recordings that are 50+ years old. A moment or two in and I don't notice anymore. I do have trouble with poorly encoded MP3s, but there's still a range where I can tell the difference and just don't care.
Also, my impression is that it would be a lot more difficult to rip Blu-ray discs to my media box, and it would certainly take up more space. So I'll probably buy a Blu-ray player at some point, perhaps for my computer (if it turns out that ripping problem is solved). But it's not worth even $200 to me right now. DVD is good enough for my needs right now.
I see two main possibilities, and not too many shades between:
(1) You want to make money off the idea, in which case you have to do what everyone else who's ever had an idea for a business has had to do: invest. Invest your time, invest your money, take the usual risks. If it's the kind of idea a competent programmer could polish off in a few months even, you're not talking about a huge investment.
(2) If you've given up on profit, tell as many people as possible what your idea is. If it's a great idea, someone will pick up on it. Or more likely, someone will tell you where to download it. Unfortunately, unless you do this in the next 24 hours, you've already missed out on your chance to tell a big chunk of Slashdot's readers, which would've been a good forum. People who read Slashdot, almost by definition, have a lot of time to waste.
Most of the other responses to this thread have been skeptical, and I am too. Chances are slim whatever you're contemplating doesn't exist, could be developed easily, and would make you any meaningful money if you exploited it commercially. Sight unseen, chances are also slim it would get to be widely known even if you released it tomorrow as a GPLed, bug-free, cross-platform work of art. But if it's really something you want to develop, and you can't do it yourself, the choices seem pretty obvious: keep it secret and invest your own time/money, or tell everyone and give up the opportunity to make it a business.
Re:it is called metonymy
on
Subject to Change
·
· Score: 4, Insightful
Happy to help. The reason is illiteracy. Figurative speech and writing is quite common, and people who insist on seeing only the "literal" meaning (a nebulous concept to anyone who has studied language from any perspective) are basically illiterate. Roll your eyes all you like, the quoted material contained a perfectly fine (and very pedestrian) use of figurative writing, and the reviewer inexplicably tried to read it literally. I personally think it was just a bit of brainlock, but I could be wrong.
Strange thing to me is that the amount of (wasted) heat energy is huge as compared to the amount of light emitted; yet these LEDs lamps are considered energy efficient.
"smellsofbikes" pointed out that incandescents radiate a lot of IR, so a lot of the wasted energy goes unnoticed. LEDs do waste a lot less energy, but a ridiculous proportion of the energy they waste is in the form of heat right at the socket. As I understand it now.
The ones I've seen don't get hot, but then again they don't generate much light either. I don't understand the distinction you're making. Are you saying the bulb doesn't get hot, just the end that goes in the socket? I guess that's a decent alternative to having the exposed part of the bulb get hot. Aside from what's heated how much, how does it compare to incandescents in terms of the amount of heat generated? It seems like if you can get the same amount of light with less wattage, you shouldn't be able to generate as much heat, give or take.
LED light bulbs are coming along. I've been keeping my eye on the GeoBulb sold by C Crane, but it's about $120 and always seems to be listed as out of stock. It is available in three color temperatures, however, so perhaps that's been worked out. There are a few others, including some T8 replacements. Unfortunately, they're all expensive and the ones that fit standard sockets max out at about the equivalent of a 60 watt incandescent. But these things will get worked out. On the plus side, advantages include not just electrical savings and longevity, but also less heat (less risk of fire) and less vulnerability to things like vibration and moderate temperature changes. Unlike compact fluorescents, they contain no mercury and turn on/off instantly. I believe some are dimmable, but I'm not sure. In any case, it seems clear that it will only take a few years of fairly routine development for LED bulbs to be much more practical than all the alternatives. There doesn't seem to be a down side.
The summary would seem to imply only modest electrical savings. I'm pretty sure the 10% figure just reflects the fact that light bulbs are only part of your electrical picture. The rated wattage for a 60-watt equivalent is about 8 watts. Correcting for overoptimism, that's about 80% savings wherever you plug one of these in.
There's no question that when you rely on someone else's code, whether it's free or commercial, you aren't guaranteed anything about future versions. They may be released under licenses you can't abide, they may not work the way you want them to, they may suddenly become unaffordable, they may stagnate and become obsolete. But as you say, writing everything from scratch is generally worse, so you take your chances.
But if the question is whether the GPLv3 is irrelevant because the kernel hasn't switched to it, the answer is clearly no. Many libraries have switched to GPLv3, and the easiest course for downstream projects will be to do the same, regardless of how they feel about the GPLv3.
Is the GPLv3 even meaningful if the kernel does not change licenses?
The incompatibility of GPLv2 and GPLv3 makes it a little viral. I have a fairly small open source project, but we depend on three libraries that have gone with GPL3. I didn't really want to switch, but if I want to use the latest versions of those libraries, I have no choice.
I suspect that the number of projects going with GPLv3 would be greater if people met their legal obligations.
But from what I have read about this issue, 'stealing' ID often leads to more creative output and profitability overall. one persons loss is a gain for many others. There's no question that the overextension of IP law is harmful. One obvious example of this is fan fiction (I think that's the term) -- derivative works written by fans (e.g., if you wrote your own Star Wars novel to post on your web site). I don't have the link handy, but a few years ago, some lawyer wrote a very cogent article on the subject, providing a pretty good argument that holders of IP often enforce their rights in a way that is destructive to everyone, giving examples from countries where lawyers do not, as a matter of culture, sue fans for helping promote their clients' works. Certainly IP law is overextended enough without add-ons for more draconian enforcement.
But it's still a little hard to see that there shouldn't still be some role for IP law. Right now, IP theft costs some people some money. We don't know exactly how much. But if there were no IP law, many of the activities from which law-abiding people like myself refrain would be much more commonplace, "costing" holders of IP much more money compared to a world with perfectly enforced IP law. So sure, the music industry needs to adapt better. I wouldn't mind it if most of the "industry" part vanished. But don't you think if redistributing the music you buy via bittorrent were legal, there would be many fewer musicians? Publishing something would be roughly equivalent to making it available for free, and everyone has to eat. I've written a book, and I didn't expect to make much if any profit on it. But my life is more complicated now, and it would be hard to justify doing it again if I knew everyone could just download it (especially if we fast-forward to a world with better ebook readers).
So I'm not saying IP law isn't broken and mis-enforced, but I really believe that the laws increase the financial incentive to create stuff, and that this has some real benefits. Nobody really knows what it would be like if there were no IP laws, but I'm pretty sure I wouldn't have to buy music anymore, or software, or certain kinds of books. To the extent these laws are unenforceable, that's really the world we live in. More so every day. So it's not that hard to see it as a problem, even if you don't like the solutions the RIAA, MPAA, and US government have been toying with (I sure don't). Remember, the existing laws weren't put into place to criminalize file sharers. The US government just happens to be looking into new (and very evil) methods of enforcement.
Re: $3 movies in the far east, I believe it's reasonably well established that imported products (software, movies) don't do nearly as well as they should, even if they're still profitable. But I can't point to any unbiased studies.
Re: me, i've personally given away two things that i could have chosen not to: a software package for brain image analysis and various writings on poker (less of the latter than i'd like, my web site has been pretty dead since poker took off). other bits of software and writing in the past as well.
I get your point: you have a basic understanding of sarcasm. However, let me state my point again: there are many cases in which theft (or whatever you want to call it) of IP costs people money. Quite obviously this is not true in all cases, and clearly not in some of the obvious cases you mentioned. But if you really believe that theft (or whatever you want to call it) of IP never ever ever ever costs anyone money, then you need to do better than a small handful of examples. And you'd better have a story that covers China.
Again, just to be ridiculously clear, if you want to contradict what I said, you need to argue that IP theft never costs the owner anything (as the original poster stated), not just that it sometimes costs the owner nothing.
Lastly, please do not use phrases like "you are right" to preface a complete mischaracterization of my views on subjects I have never discussed with you. Otherwise I'll be forced to say things like: "you are right, books and DVDs are made entirely from discarded fruit rinds." If you think your phrase actually characterizes my views, then I assume you have already done your research and can provide a plausible account for why I've given away some of my own IP for free.
While the term "intellectual property" doesn't have an upstanding motivation behind it, you should get over it. Intellectual property is now a term that has meaning, and if you can't understand that "property" doesn't always mean exactly the same thing in every possible context, then you will have a hard time understanding virtually any sentence in English. There are many very serious and disturbing problems with this kind of legislation. But the use of the term "intellectual property" is not one of them.
That said, there are many cases in which stealing an idea costs the original owner something. If you can't think of any, then you owe it to yourself to familiarize yourself with the music, publishing, software, and movie industries, to name a few.
If I want to have a book, right now that means I want it on paper, because electronic readers aren't good enough for me (I don't have one anyway). Photocopying is not a good option, although I've seen it done. So you pretty much have to buy the book the book or take it out of the library. But fast forward to a world in which electronic readers are prevalent. It's almost impossible to believe that illicit sharing wouldn't be rampant.
The only reason e-book piracy isn't rampant is because e-book use isn't rampant.
If you really believe that all IP is based on this, then your scope of experience is ridiculously narrow. Let me give two obvious counter-examples. First, developers who have release code under the GPL do so with explicit goals that are exactly the opposite of what you describe. Second, authors typically like to control the rights to publish their works, because this makes their work more marketable, and makes it possible for them to eat without having too many additional jobs. Both of these groups of people rely on intellectual property laws to accomplish these goals, and without IP laws of some kind, there would be no GPLed software, and perhaps a tenth the number of authors we have today.
Everyone around here seems to hate tradition for tradition's sake unless it's a computer related tradition. Congratulations, you've become what you hate and you didn't even realize it. That's because consistent use of units of measurements is not tradition for tradition's sake, it's so that we can tell how much stuff will fit on a storage device. As luck would have it, we have a decent system for that, and it's based on round numbers in base 2. I don't think the system has any real drawbacks, although apparently you find it somehow clumsy. But as soon as you introduce a second system, suddenly it becomes difficult to tell how many bytes a 120GB drive will hold, unless you happen to keep track of exactly when each manufacturer switched over, or in general to measure things precisely. Probably if the entire industry agreed to switch all at once, it could be settled in about ten years. But why bother? Nobody really cares if a gigabyte is this size or that size, the only thing we really care about is that it always means the same thing. And this really has nothing to do with computers, but fortunately the companies that distribute orange juice and things like that haven't yet figured out a way to give me less (while still maintaining some asinine argument about why it's still a "quart"). Actually, for all I know they have. The size of a quart is much more of an arbitrary tradition than the size of a gigabyte. If I find out they've been doing that, I'll be a little irked. Not because of tradition, but because I don't like being cheated.
You're not a fool for seeing value in the things you love and wanting to reward artists for their dedication. You're a fool for thinking the RIAA somehow promotes this process. On the whole, the RIAA is deeply harmful to artists and consumers alike. If you really cared about the arts, beyond picking up the occasional Britney Spears album, then you would despise the RIAA too.
This may or may not be stupid research, but it's far from implausible. Despite the sensational headline and the fact that the authors got their article in a good place, this kind of thing (brain activity predicting future behavior) is commonplace in functional brain imaging. I'm not saying your "I don't believe it" critique isn't well thought out, but perhaps you would like to be a little more specific about where you think the authors have gone wrong.
As an aside, I don't know much about trees and leaves, but I think if we put a small tree in an MRI scanner, I could find some information in the image that would help me do better than chance at predicting which leaf would fall next.
Technically, all comparisons are significantly different from 0, so the probability that voxels passing any arbitrary criterion is 100% given sufficiently precise measuring devices. That's certainly true (although I'd replace "significantly" with "truly"), and I do think it's fair to say that a small minority of fMRI people understand this.
are you not familiar with the hundreds upon hundreds of articles dating from the 1950s to today explaining why statistical testing is unscientific and harmful to psychology? I've read some but not all of the articles listed on that page. I'm of course familiar with the arguments generally, and I of course have many concerns about significance testing, even though I often use it (sometimes for good reasons, sometimes for bad). You should note that on the very web page you cited, the title and summary both suggest an indictment of "indiscriminate" use of significance testing. I suspect few of the authors cited would argue that significance testing is in all cases harmful. Certainly there is lots of indiscriminate significance testing in fMRI, and plenty of bad statistical practice in general. Often out of ignorance, other times mostly to satisfy reviewers. I don't want to comment on the specific article you mentioned, but I will note that while I'm sure Science is in many ways a fine journal, in the arena of fMRI, they typically publish the flashiest research, not the best executed. Voxel-by-voxel significance testing is sometimes useful and sometimes not. While widely used, it's only one of many approaches used with fMRI.
All that said, this is a completely orthogonal issue. For the sake of argument only, I will pretend to believe that significance testing is universally a bad idea, and that every fMRI study ever published has been a load of nonsense. It still remains false that voxels meeting the Bonferroni-corrected statistical criterion in typical size brain volumes are unlikely to exist. Whether or not SPM'ing is a good idea, and whether or not this is a meaningful way to look at a given dataset, this is very easily refuted.
Now, I'm sorry to hear that the APS promotes bad science and/or statistical practice. That's an indictment of the APS, not of researchers who use fMRI. I frankly don't even know anyone who's an APS member, although I do still remember some of the unintentionally humorous junk mail they sent me when I was in grad school. I'd be surprised if more than a tiny percentage of fMRI users who would even be aware of APS pronouncements. So while again, I don't want to defend the entirety of the fMRI community, it would be false to say no one in the fMRI community cares about these things, and particularly pointless to support this point by reference to the APS.
You greatly overestimate the extent to which psychologists care how much they are wasting grant money conducting bad research. I'm surprised you would venture an opinion on this, since I've never said anything publicly about how much I think psychologists care about wasting grant money. I do happen to have many data points on this in my personal experience, and it's quite depressing. I don't think there's room for my view to be more than a moderate overestimate, and I do know some pretty slimy people.
There's no question that your story about a researcher with no clue what s/he was doing is repeated often in psychology, and probably in other fields as well.
However, your example from fMRI speaks to complete ignorance of the field, and I'd like to force you to defend it. Thousands of fMRI experiments have been carried out, and this standard for significance is often met. When you say "very unlikely to actually exist," I can't imagine what you're thinking, since this statement is so easily falsified (in fact, in many publicly available datasets). Although not everyone does power analysis well, it's no secret how large an effect can be detected with fMRI, as a function of a few dozen easily estimated variables. Detectable effects are in fact very common. Your example seems to assume Bonferroni correction, which is generally not optimal. But even if we assume that it is, many many experiments produce results that exceed these thresholds, even though they do mean you can't study subtle effects in small groups.
If you want to defend your view that fMRI users don't grasp that they don't grasp it (I'm sure I would be included in your blanket statement), then you are doing a profound disservice to science by not submitting your work to NeuroImage (or if you think you wouldn't get a fair review there, there are certainly other journals you should consider). I want to rephrase myself here, to be perfectly clear. If you really do know of a flaw in statistical practice that affects many thousands of studies and many millions (probably billions) of dollars of grant money, but that has escaped the notice of everyone in the field, and you haven't taken the time to submit your insight to a decent journal, then you are the worst kind of bad scientist.
Of course it's not secret that many researchers do the statistics poorly (even making allowances for what's practical), and it's inexcusable. It's often for the same reason you mentioned: people just do what they saw in some other article, without understanding it. And many such studies are underpowered, although that's a problematic issue in itself. I would also agree with the statement that the SPM approach to fMRI has some very worrisome weaknesses, although not the one you identified. However, I will say that if you want to make such sweeping claims about what is and isn't nonsense, in a field in which you have obviously only dabbled (if that), you should make the basis for your criticism clearer, as naive readers could easily get the wrong idea.
Amazon is currently, to paraphrase Winston Churchill, the worst possible online bookstore, except for all the alternatives. From a customer point of view, the web site is comically bad. From a small publisher point of view, they seem at best hostile. They're good for many things, and I'll continue to use them, but one-stop shopping for books is no longer something they're willing to provide. Yes, this will hurt certain publishers, in the same way many publishers are hurt when Barnes & Noble, Wal-Mart, and the local grocery store won't stock their books. But for the moment, they're free to sell what they want. Small publishers know that certain outlets are unavailable to them, so this is really nothing new.
Slashdot Mirror
I agree with most of this, and certainly there is a lot to be learned from quick and dirty analyses, simple visualizations, etc. But one of the things that really concerns me about promoting the use of Excel for analysis (as I would imagine this book does) is that it lures researchers away from tools that are more appropriate for the job, but are more difficult to learn and use. Even though it will take people more time to reach the same level of fluency with R that they would have with Excel, I think it's time well spent. So while I appreciate the uses of quick and dirty analyses as much as the next guy, I'd still rather see them done in R than in Excel. Incidentally, I know that R is a little impenetrable to some users due to the lack of a comprehensive standard graphical interface. I only use it as an example because it's what I use, and I know it has the respect of serious stats jocks. Also, I think that anyone who's smart enough to know what kind of analysis they need should be able to learn R. Of course, for many of the more involved problems in many fields, you need software designed specifically for the job anyway, and neither R nor Excel nor any other general-purpose package will do.
On a slight tangent, one of the reasons I like openoffice (I ditched Word years ago, and the new release candidate is really pretty good) is that I try to avoid using it for things that neither it nor the Microsoft alternatives are good for. Charting is a great example, so I'm with you on that. I'm abysmal at charting in R, but it does tend to have what I need.
Care to expand on why you think you can't do 'meaningful scientific data analysis in Excel?' Are you one of these people who 'reviews' books without actually reading them?
Someone else has already posted a link to a page that nicely summarizes many (not all) of the problems with using excel for science. But there is virtually no statistical technique which isn't already better implemented in R (free) and many other statistical packages. Real stats packages provide implementations of a given technique that are at least as reliable, provide more control, more options, more diagnostics, and often more guidance. The built-in stuff in Excel is so oversimplified that I think if you're really forced to use it for serious statistics, you'd have to re-implement things using basic arithmetic operations. It's graphing capabilities, last time I checked, lacked the majority of even the most basic kinds of statistical/scientific graphs/plots. Sure, you can do this or that in Excel, and if you're willing to put in enough work you can often get what you really need out of it. But it's rarely if ever the best tool for the job of scientific data analysis.
I don't review books about reading them, and this is no exception. But I do have an informed opinion about the premise of this book (and to a lesser extent about the level of insight of a reviewer who seems, to put it mildly, easily awed). The premise that Excel is good to use for scientific data analysis is pretty deeply misguided. I'd be happy to be convinced otherwise if I were really wrong, but I can only set aside so much time for listening to arguments from nutcases (just in case one of them may have a point). I'm sure if I actually read this book, I'd learn about various useful things Excel can do that would surprise and impress me. But I already have all the information I need to form a reliable opinion on this question, and I value my time too much to read books about space aliens living among us or about doing analysis in Excel.
Truth to tell, I use openoffice calc (more or less an Excel clone) quite a bit for research-related things. But I'm careful with it, and don't rely on it for much more than moving numbers around.
There's something hard to reconcile about the reviewer's obvious awe and the fact that the book was written by someone who thinks doing meaningful scientific data analysis in Excel is a good idea.
Companies can suck and still produce products you'd like to use. I don't particularly like Apple (or Microsoft), but I do have reasons to want to run OSX (and Windows). One of the reasons I don't like Apple is because they make it difficult for me to support their platform, specifically by making it difficult to run on third-party or (more important to me) virtual hardware. I understand why, but understanding doesn't make me like them. There are lots of other reasons to dislike Apple, of course (e.g., their promotion of DRM-encrusted music), and very few of them are inconsistent with wanting to run OSX on the hardware of your choice. So for me, Apple does suck, and I'd still be pretty happy if I could run their OS more conveniently.
I just don't understand it when people say DVD is "good enough". You can see the compression artifacts! (and that's on a low resolution display)
DVD is absolutely good enough for me. All else being equal, of course I'd like better resolution. Sometimes I do find DVD artifacts bothersome, and I certainly resent it when they're poorly encoded. Of course, Blu-ray discs can be poorly encoded too, but it's liable to be less of a problem. Still, with my eyesight barely at 20/20 anymore, and my aging 32 inch LCD subtending no more than about 19 degrees of visual angle (that's at about 8 feet), the difference in resolution makes a lot less difference to me than to someone half my age watching a 50 inch set at the same distance. On top of which, when I watch movies, I tend to get caught up in things that are somewhat peripheral to the quality of the medium -- plot, acting, cinematography, etc. I've often watched movies with scratched film projected by crappy projectors. I often listen to audio recordings that are 50+ years old. A moment or two in and I don't notice anymore. I do have trouble with poorly encoded MP3s, but there's still a range where I can tell the difference and just don't care.
Also, my impression is that it would be a lot more difficult to rip Blu-ray discs to my media box, and it would certainly take up more space. So I'll probably buy a Blu-ray player at some point, perhaps for my computer (if it turns out that ripping problem is solved). But it's not worth even $200 to me right now. DVD is good enough for my needs right now.
I see two main possibilities, and not too many shades between:
(1) You want to make money off the idea, in which case you have to do what everyone else who's ever had an idea for a business has had to do: invest. Invest your time, invest your money, take the usual risks. If it's the kind of idea a competent programmer could polish off in a few months even, you're not talking about a huge investment.
(2) If you've given up on profit, tell as many people as possible what your idea is. If it's a great idea, someone will pick up on it. Or more likely, someone will tell you where to download it. Unfortunately, unless you do this in the next 24 hours, you've already missed out on your chance to tell a big chunk of Slashdot's readers, which would've been a good forum. People who read Slashdot, almost by definition, have a lot of time to waste.
Most of the other responses to this thread have been skeptical, and I am too. Chances are slim whatever you're contemplating doesn't exist, could be developed easily, and would make you any meaningful money if you exploited it commercially. Sight unseen, chances are also slim it would get to be widely known even if you released it tomorrow as a GPLed, bug-free, cross-platform work of art. But if it's really something you want to develop, and you can't do it yourself, the choices seem pretty obvious: keep it secret and invest your own time/money, or tell everyone and give up the opportunity to make it a business.
Happy to help. The reason is illiteracy. Figurative speech and writing is quite common, and people who insist on seeing only the "literal" meaning (a nebulous concept to anyone who has studied language from any perspective) are basically illiterate. Roll your eyes all you like, the quoted material contained a perfectly fine (and very pedestrian) use of figurative writing, and the reviewer inexplicably tried to read it literally. I personally think it was just a bit of brainlock, but I could be wrong.
Strange thing to me is that the amount of (wasted) heat energy is huge as compared to the amount of light emitted; yet these LEDs lamps are considered energy efficient.
"smellsofbikes" pointed out that incandescents radiate a lot of IR, so a lot of the wasted energy goes unnoticed. LEDs do waste a lot less energy, but a ridiculous proportion of the energy they waste is in the form of heat right at the socket. As I understand it now.
The ones I've seen don't get hot, but then again they don't generate much light either. I don't understand the distinction you're making. Are you saying the bulb doesn't get hot, just the end that goes in the socket? I guess that's a decent alternative to having the exposed part of the bulb get hot. Aside from what's heated how much, how does it compare to incandescents in terms of the amount of heat generated? It seems like if you can get the same amount of light with less wattage, you shouldn't be able to generate as much heat, give or take.
LED light bulbs are coming along. I've been keeping my eye on the GeoBulb sold by C Crane, but it's about $120 and always seems to be listed as out of stock. It is available in three color temperatures, however, so perhaps that's been worked out. There are a few others, including some T8 replacements. Unfortunately, they're all expensive and the ones that fit standard sockets max out at about the equivalent of a 60 watt incandescent. But these things will get worked out. On the plus side, advantages include not just electrical savings and longevity, but also less heat (less risk of fire) and less vulnerability to things like vibration and moderate temperature changes. Unlike compact fluorescents, they contain no mercury and turn on/off instantly. I believe some are dimmable, but I'm not sure. In any case, it seems clear that it will only take a few years of fairly routine development for LED bulbs to be much more practical than all the alternatives. There doesn't seem to be a down side.
The summary would seem to imply only modest electrical savings. I'm pretty sure the 10% figure just reflects the fact that light bulbs are only part of your electrical picture. The rated wattage for a 60-watt equivalent is about 8 watts. Correcting for overoptimism, that's about 80% savings wherever you plug one of these in.
Optimizing for boot time over everything else seems very foolish to me.
I guess that's true if you're designing a web server. Probably not if you're designing a computer-controlled defibrillator.
There's no question that when you rely on someone else's code, whether it's free or commercial, you aren't guaranteed anything about future versions. They may be released under licenses you can't abide, they may not work the way you want them to, they may suddenly become unaffordable, they may stagnate and become obsolete. But as you say, writing everything from scratch is generally worse, so you take your chances.
But if the question is whether the GPLv3 is irrelevant because the kernel hasn't switched to it, the answer is clearly no. Many libraries have switched to GPLv3, and the easiest course for downstream projects will be to do the same, regardless of how they feel about the GPLv3.
Is the GPLv3 even meaningful if the kernel does not change licenses?
The incompatibility of GPLv2 and GPLv3 makes it a little viral. I have a fairly small open source project, but we depend on three libraries that have gone with GPL3. I didn't really want to switch, but if I want to use the latest versions of those libraries, I have no choice.
I suspect that the number of projects going with GPLv3 would be greater if people met their legal obligations.
dan
But it's still a little hard to see that there shouldn't still be some role for IP law. Right now, IP theft costs some people some money. We don't know exactly how much. But if there were no IP law, many of the activities from which law-abiding people like myself refrain would be much more commonplace, "costing" holders of IP much more money compared to a world with perfectly enforced IP law. So sure, the music industry needs to adapt better. I wouldn't mind it if most of the "industry" part vanished. But don't you think if redistributing the music you buy via bittorrent were legal, there would be many fewer musicians? Publishing something would be roughly equivalent to making it available for free, and everyone has to eat. I've written a book, and I didn't expect to make much if any profit on it. But my life is more complicated now, and it would be hard to justify doing it again if I knew everyone could just download it (especially if we fast-forward to a world with better ebook readers).
So I'm not saying IP law isn't broken and mis-enforced, but I really believe that the laws increase the financial incentive to create stuff, and that this has some real benefits. Nobody really knows what it would be like if there were no IP laws, but I'm pretty sure I wouldn't have to buy music anymore, or software, or certain kinds of books. To the extent these laws are unenforceable, that's really the world we live in. More so every day. So it's not that hard to see it as a problem, even if you don't like the solutions the RIAA, MPAA, and US government have been toying with (I sure don't). Remember, the existing laws weren't put into place to criminalize file sharers. The US government just happens to be looking into new (and very evil) methods of enforcement.
Re: $3 movies in the far east, I believe it's reasonably well established that imported products (software, movies) don't do nearly as well as they should, even if they're still profitable. But I can't point to any unbiased studies.
Re: me, i've personally given away two things that i could have chosen not to: a software package for brain image analysis and various writings on poker (less of the latter than i'd like, my web site has been pretty dead since poker took off). other bits of software and writing in the past as well.
I get your point: you have a basic understanding of sarcasm. However, let me state my point again: there are many cases in which theft (or whatever you want to call it) of IP costs people money. Quite obviously this is not true in all cases, and clearly not in some of the obvious cases you mentioned. But if you really believe that theft (or whatever you want to call it) of IP never ever ever ever costs anyone money, then you need to do better than a small handful of examples. And you'd better have a story that covers China.
Again, just to be ridiculously clear, if you want to contradict what I said, you need to argue that IP theft never costs the owner anything (as the original poster stated), not just that it sometimes costs the owner nothing.
Lastly, please do not use phrases like "you are right" to preface a complete mischaracterization of my views on subjects I have never discussed with you. Otherwise I'll be forced to say things like: "you are right, books and DVDs are made entirely from discarded fruit rinds." If you think your phrase actually characterizes my views, then I assume you have already done your research and can provide a plausible account for why I've given away some of my own IP for free.
While the term "intellectual property" doesn't have an upstanding motivation behind it, you should get over it. Intellectual property is now a term that has meaning, and if you can't understand that "property" doesn't always mean exactly the same thing in every possible context, then you will have a hard time understanding virtually any sentence in English. There are many very serious and disturbing problems with this kind of legislation. But the use of the term "intellectual property" is not one of them.
That said, there are many cases in which stealing an idea costs the original owner something. If you can't think of any, then you owe it to yourself to familiarize yourself with the music, publishing, software, and movie industries, to name a few.
If I want to have a book, right now that means I want it on paper, because electronic readers aren't good enough for me (I don't have one anyway). Photocopying is not a good option, although I've seen it done. So you pretty much have to buy the book the book or take it out of the library. But fast forward to a world in which electronic readers are prevalent. It's almost impossible to believe that illicit sharing wouldn't be rampant.
The only reason e-book piracy isn't rampant is because e-book use isn't rampant.
If you really believe that all IP is based on this, then your scope of experience is ridiculously narrow. Let me give two obvious counter-examples. First, developers who have release code under the GPL do so with explicit goals that are exactly the opposite of what you describe. Second, authors typically like to control the rights to publish their works, because this makes their work more marketable, and makes it possible for them to eat without having too many additional jobs. Both of these groups of people rely on intellectual property laws to accomplish these goals, and without IP laws of some kind, there would be no GPLed software, and perhaps a tenth the number of authors we have today.
You're not a fool for seeing value in the things you love and wanting to reward artists for their dedication. You're a fool for thinking the RIAA somehow promotes this process. On the whole, the RIAA is deeply harmful to artists and consumers alike. If you really cared about the arts, beyond picking up the occasional Britney Spears album, then you would despise the RIAA too.
This may or may not be stupid research, but it's far from implausible. Despite the sensational headline and the fact that the authors got their article in a good place, this kind of thing (brain activity predicting future behavior) is commonplace in functional brain imaging. I'm not saying your "I don't believe it" critique isn't well thought out, but perhaps you would like to be a little more specific about where you think the authors have gone wrong.
As an aside, I don't know much about trees and leaves, but I think if we put a small tree in an MRI scanner, I could find some information in the image that would help me do better than chance at predicting which leaf would fall next.
All that said, this is a completely orthogonal issue. For the sake of argument only, I will pretend to believe that significance testing is universally a bad idea, and that every fMRI study ever published has been a load of nonsense. It still remains false that voxels meeting the Bonferroni-corrected statistical criterion in typical size brain volumes are unlikely to exist. Whether or not SPM'ing is a good idea, and whether or not this is a meaningful way to look at a given dataset, this is very easily refuted.
Now, I'm sorry to hear that the APS promotes bad science and/or statistical practice. That's an indictment of the APS, not of researchers who use fMRI. I frankly don't even know anyone who's an APS member, although I do still remember some of the unintentionally humorous junk mail they sent me when I was in grad school. I'd be surprised if more than a tiny percentage of fMRI users who would even be aware of APS pronouncements. So while again, I don't want to defend the entirety of the fMRI community, it would be false to say no one in the fMRI community cares about these things, and particularly pointless to support this point by reference to the APS. You greatly overestimate the extent to which psychologists care how much they are wasting grant money conducting bad research. I'm surprised you would venture an opinion on this, since I've never said anything publicly about how much I think psychologists care about wasting grant money. I do happen to have many data points on this in my personal experience, and it's quite depressing. I don't think there's room for my view to be more than a moderate overestimate, and I do know some pretty slimy people.
There's no question that your story about a researcher with no clue what s/he was doing is repeated often in psychology, and probably in other fields as well.
However, your example from fMRI speaks to complete ignorance of the field, and I'd like to force you to defend it. Thousands of fMRI experiments have been carried out, and this standard for significance is often met. When you say "very unlikely to actually exist," I can't imagine what you're thinking, since this statement is so easily falsified (in fact, in many publicly available datasets). Although not everyone does power analysis well, it's no secret how large an effect can be detected with fMRI, as a function of a few dozen easily estimated variables. Detectable effects are in fact very common. Your example seems to assume Bonferroni correction, which is generally not optimal. But even if we assume that it is, many many experiments produce results that exceed these thresholds, even though they do mean you can't study subtle effects in small groups.
If you want to defend your view that fMRI users don't grasp that they don't grasp it (I'm sure I would be included in your blanket statement), then you are doing a profound disservice to science by not submitting your work to NeuroImage (or if you think you wouldn't get a fair review there, there are certainly other journals you should consider). I want to rephrase myself here, to be perfectly clear. If you really do know of a flaw in statistical practice that affects many thousands of studies and many millions (probably billions) of dollars of grant money, but that has escaped the notice of everyone in the field, and you haven't taken the time to submit your insight to a decent journal, then you are the worst kind of bad scientist.
Of course it's not secret that many researchers do the statistics poorly (even making allowances for what's practical), and it's inexcusable. It's often for the same reason you mentioned: people just do what they saw in some other article, without understanding it. And many such studies are underpowered, although that's a problematic issue in itself. I would also agree with the statement that the SPM approach to fMRI has some very worrisome weaknesses, although not the one you identified. However, I will say that if you want to make such sweeping claims about what is and isn't nonsense, in a field in which you have obviously only dabbled (if that), you should make the basis for your criticism clearer, as naive readers could easily get the wrong idea.
Amazon is currently, to paraphrase Winston Churchill, the worst possible online bookstore, except for all the alternatives. From a customer point of view, the web site is comically bad. From a small publisher point of view, they seem at best hostile. They're good for many things, and I'll continue to use them, but one-stop shopping for books is no longer something they're willing to provide. Yes, this will hurt certain publishers, in the same way many publishers are hurt when Barnes & Noble, Wal-Mart, and the local grocery store won't stock their books. But for the moment, they're free to sell what they want. Small publishers know that certain outlets are unavailable to them, so this is really nothing new.