...which is that the authors are basing their conclusions on the premise that physiological responses are purely indicative of specific emotional states. This is a position people like Ekman take as well, and it's easily falsified. Russell reviewed the literature around 2000 or so and found that, in fact, emotional displays are at least somewhat socially motivated and don't always equate to specific emotions in a one-to-one fashion (especially in speech). To put it more simply, you may smile because you're happy, but you might also smile because you're being sarcastic, because you're covering up frustration, etc. However, this study rests on the assumption that people only smile because they're happy -- period. (In all fairness, they're not exactly talking about smiling here, but the principle is the same.)
So, sure, if you're willing to accept that premise, then this study is great. If not, it's just another in a long line of studies that suggest, but do not convincingly prove, what emotions can be generated by particular events. Really, this study just again points out how insanely difficult it is to get to a "ground truth" of what emotions people experience.
And yes, I've read the article -- I happen to have electronic access to the journal.
Luckily, if you happen to have a copy of Office, you can go to Help->About and click on the conveniently located link to read the EULA.
Upon reading the EULA for Office 2003, I can't find the clause you mention. It's a pretty standard EULA and is actually surprisingly generous in its licensing terms, especially with regards to clip art (which is more or less "don't use it to violate any depicted entity's trademark, and if you do, we're not liable"). There's nothing in there that I could find about not using Office to hurt Microsoft. Perhaps it's in the French translation?
Actually, record companies advance the band up front, which isn't quite the same as paying them. The advance is paid out of the royalties of the album sales, so in essence, the record company is just giving them a loan. The record companies try to recoup 100% of all expenses out of royalties; this includes music videos, tours, production costs, many forms of advertising, and more. Even the band's producer is paid by the band out of a 100% recouped advance. About the only things that aren't 100% recoupable are actual pressing costs and distribution. Music publishing, you see, has very little to do with most other forms of publishing.
I recommend Donald Passman's "All You Need To Know About the Music Business" for a good overview of what record companies actually do, what the average royalties per CD sold actually are, and how recoupable advances can drive popular bands into bankruptcy. You'll discover all sorts of fun tidbits, like the 20% breakage fee on royalties (a holdover from the days of vinyl that bands are stuck with now). No, I'm not shilling -- I've read it, and it's quite enlightening.
To get back to the overriding question, the answer is almost assuredly yes. Radiohead most likely made more money off the download sales than they would have off a physical CD sale, since their royalties per CD are likely less than $3.
The financial aid departments at all universities always engage in specialized calculations for financial aid. While they may offer truly impressive packages to smart people, there's a point at which they're going to insist that someone other than them and the federal government pay. That someone is going to either be you or your parents, and if the coursework is difficult enough that a job is out of the question, and your parents won't front the cash... you're out of luck. My problem wasn't being accepted; my problem was paying for it. This is far more common than you seem to believe. I speak from not only personal experience, but also for several people I know who were accepted to places like Harvard and Stanford with scholarships and federal aid, but couldn't make up the difference.
I'll also point out that, as someone who does hire undergrads for research on grants, we *always* make them work for credit for a semester or two first before putting up some money, and we pay rather poorly ($6.50 an hour, generally, max of 20 hours per week, no benefits). Try making up a few thousand a year in tuition and pay for books and pay for food on that.
The problem is not so much that people are doing research in this field -- people still do research into parapsychology and memetics, for example. The problem is asserting that your theoretical framework is true and correct in the face of serious competition and disconfirmatory evidence. Homeopathy's principle claims are not supported by evidence. As a theoretical framework, it doesn't buy us anything in terms of explanatory power over its primary competitor, the placebo effect. The placebo effect is even more predictive, because it can explain results such as "red and purple liquids, colored by a biologically non-reactive dye, have greater treatment effects than clear ones." How does homeopathy address that? Even clinically, homeopathy fails; its results are on par with what you'd predict from placebo.
I don't mind if people spend time looking for results they may never find. It's true that they might stumble upon something, though the evidence so far suggests that they most likely won't. Given the results thus far, we should definitely consider research into homeopathy very risky, and be mindful of spending money on it. That's an issue of efficient resource allocation, however.
My major problems with researchers into homeopathy is that they often violate the epistemological underpinnings and conventions of science (no special pleading, peer review of results, full disclosure of methods, falsifiable theories and hypotheses, etc.), and that they often make assertions that go far beyond, or run completely counter to, the results of their studies. Those two problems cut to the core of why it's a pseudoscience: it claims to be a science, and sometimes even puts on the airs and trappings of scientific pursuits, but it doesn't follow the same epistemological rules and therefore is *not* science.
The Giant Phone Bills of Doom seem to be iPhone specific. I have an unlimited data plan from AT&T, I use it with my HTC Hermes, and my phone bill has never exceeded 25 pages. And that's with three different phone lines on the bill.:)
And yes, before you ask, the AT&T billing system does do detailed data billing for me. Perhaps it's something different in the way the iPhone interfaces with the GSM network, or perhaps iPhone data plan users get different billing procedures.
I can definitively say, however, that any smartphone should have its data connection turned off overseas, or especially on a cruise. Roaming rates on cruise ships are well and truly insane. Roaming data rates in general are disturbingly high.
Ekman's research is based primarily on the theoretical principle that facial emotional displays are an automatic, uncontrolled process -- in other words, that you smile reflexively because you're happy. In this paradigm, attempting to restrict the display of a facial display will produce a strange expression that'll be easily recognized as fake.
Obviously, that doesn't explain everything. Much to my chagrin, I actually do research in this messy field, and what we've found is that the Ekman approach isn't as good as one would think. One of the issues of Ekman's research is that it's typically done on staged photographs. When we use emotion raters trained on Ekman's Facial Action Coding System to judge the emotional displays in videos of real people making real displays, the inter-rater reliability falls through the floor. (It's even worse if you split off from Ekman's "basic emotion" categories, which are of dubious utility in the real world anyway.) It's not as bad as untrained raters, but it's still not great. This evidence suggests that there's something else going on besides an automatized process. Russell, another researcher in this field, purports an alternative explanation: that emotional displays are in many cases controlled social processes, and can't really be interpreted outside of a social context.
In any event, the parent is right. The claims on "micro-expressions" and Ekman's FACS in general are not nearly as defensible as their proponents in the TSA would like you to believe. From a signal detection standpoint, the problem isn't so much that you'll have misses, but that you'll have false alarms. More worrying, though, is our line of research shows that the miss and false alarm rates are in many ways a function of individual differences. In other words, some raters err more towards too many identifications (high FA rate), and others err towards too few (high miss rate).
As boring as it sounds, more research is needed before this is implemented -- "this" being any security measure based on Ekman's research.
MFC prevents this (though, interestingly, the.NET Framework doesn't:) ) because, internally, the GDI window code isn't thread-safe. If you call windowing functions for a window handle from a different thread than the one that owns that handle, Bad Things Happen. It's a fundamental rule of Windows GUI programming, albeit a somewhat obscure one for most users. Windows XP and Vista are substantially better at dealing with violations of the rule, but Windows 9x OSes would often go down completely if you broke that rule. Apps under XP sometimes work properly, and sometimes blow up with inscrutable exceptions.
The solution, as.NET programmers should all know, is to marshal the calls onto the appropriate thread for the control. In.NET, that means checking Control.InvokeRequired and using Control.Invoke with an appropriate delegate for the method you want to call if that property is true (i.e., you're not on the thread that created the control). I've not done very much MFC programming at all, so I'm not exactly sure how you'd handle this issue over there -- perhaps with a custom event queue, maybe?
Okay, people keep saying this... but when I report my "drug sales" net income (after amortizing my.45 and deducting bribes), won't they just turn that right around and charge you with a crime, implicitly requiring you to waive the fifth?
They can't directly charge you with a crime AFAIK, though perhaps they might be able to if you did something really boneheaded, like write "illegal drug sales" in the "type of business" section of your tax return.
However, your tax returns can be used as evidence against you in a criminal case. If the police suspect you might be selling drugs, they can acquire your tax records. If those tax records show exorbitant income for your professed job, that could be used as evidence sufficient to acquire a search warrant, and then you're boned. Certain types of transactions have to be reported and made available to law enforcement agencies as well. The question for the criminal then becomes a risk analysis: is it safer to conceal the income from the IRS, or report it and hope it's sufficiently hidden or justifiable?
This page describes some of the IRS' activities in narcotics enforcement.
Just to clarify, AT&T has basically four different data plans based on the type and -- so they say -- intended usage of the device. The cheapest one is for "smartphones," which covers all phones without keyboards. The "PDA" plan is the next most expensive, for all phones with keyboards (HTC phones, Treos, etc.) except for Blackberries, which have their own plan. The most expensive is the "computer" plan, which covers their HSPDA/EDGE PC Cards for use in laptops. Some of the plans have a tethered option, where if you pay extra, you can use your phone as a network gateway for your computer. I had heard they were planning on dispensing with the Blackberry plan and rolling it into the PDA plan, but that's second-hand from a CSR, so take it with a grain of salt.
The price of the plan is primarily based on the expected data usage. Additionally, all smartphones have to go through a WAP gateway (MediaNET) for their Internet access. It's an open secret that AT&T will hit you up with per-kb data charges or demand you upgrade to a more costly plan if you use a large amount of transfer for your type of device or if they catch you using their direct connection servers on the cheap smartphone data plan.
Unless Apple has a special deal for the iPhones, they'll be charged either per-kb or on the PDA Connect plan, which is currently $39.95/mo. for unlimited data. They *might* end up creating an iPhone specific plan on the grounds that they may not support XpressMail/GoodMail/DirectPush and thus will use less data than your average business smartphone, but I doubt it. I don't suspect AT&T will apply an additional fee to iPhone users.
The Win32 API can technically support 2^16 mouse buttons (or 2^32 on Win64) in addition to left, right, and middle. The limit is in the design of the WM_XBUTTON messages, where the high order word of the w message parameter specifies which extra button was clicked. Right now, the OS only generates events for XBUTTON1 and XBUTTON2, the two extra buttons on the Intellimouse and its clones. WM_APPCOMMAND messages are used to handle any excess buttons, which are mapped to specific application commands.
System.Windows.Forms in the Framework supports all 5 mouse buttons in its various Mouse events.
Every MMO I've played has had very strict internal oversight on its GMs and devs. They can't use their GM account to play non-GM characters. They can't play with their normal player account on servers that they have GM duties on (I'm aware that wouldn't work with EVE, though). When a "canon character" is needed for an event, there's a sign-out and approval process, and the people who are allowed access to such characters are limited. Log audits of GM abilities are performed periodically to make sure that people aren't abusing their powers.
Yes, certainly, some things slip through -- on Firiona Vie on EQ, someone I played with regularly was given a 100% CSR rez item by a guide, with the understanding that it'd only be used to rez people who died due to bugs. That said, strong internal controls would help protect everyone against the sort of malfeasance that may, or may not, be happening on EVE. This election likely won't solve anything, especially since the winners will likely be members of the large corporations/alliances, who are the people being accused of cheating in the first place.
The fact that the APIs are published is immaterial, really. I know of at least one company's software that has a similar clause in its EULA to the Visual Studio one, and they don't do even the slightest protection of their fully documented APIs. Nevertheless, using them would violate the EULA if I wasn't licensed to do so. That's the contract I entered with that company. If I want to use those APIs, I need to pay them more money for the license to do so. This isn't anything terribly new, and even "good guys" do it now and then to protect a revenue stream (cf. MySQL's multiple licenses).
For those not clear on the situation, the short of it is this. TestDriven.NET is an add-in for Visual Studio. Visual Studio Express has a "technical limitation" that ostensibly prevents the loading of add-ins (removal of the Add-In Manager, I believe). The EULA states that:
"...you may use the software only as expressly permitted in this agreement. In doing so you must comply with any technical limitations in the software that only allow you to use it in certain ways... You may not work around any technical limitations in the software."
Constructing an add-in that can be loaded by Express is presumably a violation of the EULA for Express, because you're working around the technical limitation (weak though it may be) in the software that blocks the loading of add-ins. Technically speaking, anyone who uses it with Express is also violating the EULA. The best argument, IMO (and IANAL), is going to be that disabling the Add-In Manager isn't really a technical limitation against the loading of add-ins, since they can be loaded programmatically. It's a technical limitation against end-users manually loading add-ins.;)
I regularly read and edit Word and Excel documents on my Cingular 8125. It's often much easier for me to make modifications to an e-mailed document or write a new document on it than to pull out my laptop and boot it up -- seeing as my phone is always with me, while my laptop isn't. As a graduate student, I also find it's handy for taking notes in class or in meetings so I can e-mail them later or sync them up to my computer for further editing and printing. I won't say that this is true for everyone, but I will say that I feel my productivity has gone up since getting my 8125 (to replace my Treo 600, which itself was a big leap over my old Nokia, uh, "dumbphone").
That said, I think the iPhone will fail as a corporate device not for two reasons, neither one of which is not having Office. One, it's unlikely it will support direct push e-mail on Exchange without GoodLink or some custom carrier solution (such as Cingular XpressMail), which will increase internal support costs by interfering with platform standardization. This is basically the same reason why other software and hardware monocultures appear in corporations. Second, the software lockdown aspect, while potentially able to be circumvented, will interfere with custom corporate applications, which are a common thing to put on corporate-owned phones. If nothing else, the fact that the phone runs on OS X will require new versions of these custom applications, which gets back to increasing costs.
The iPhone's neat, sure. For people out of the business environment and who have a whole lot of money to spend, I'm sure it'll be wonderful. On the other hand, businesses will keep buying the same easy to integrate devices they've been buying unless Apple can give them an extremely compelling reason to switch. So far, they haven't done that.
The methodology wasn't flawed, so much as the analysis and the conclusions drawn from it.
A PEAR experiment involved a participant attempting to influence a random number generator (essentially) in a pre-specified direction over a large number of trials. Because random events are, by nature, random, you can get streaks that are above or below the mean. If you analyze a large enough sample, these streaks can become statistically significant, even though they're essentially meaningless and practically insignificant -- it's similar to the fact that any deviation from the mean, no matter how small, is statistically significant if you measure the entire population. Additionally, while the probability of any particular streak is low (.5^n is the probability of any number of heads flipped in a row, which gets very small when you talk about enough of them), if you have enough random events, those streaks are pretty much guaranteed to appear.
So, that's the logic of the PEAR data analysis. Collect a huge corpus of random events, look for streaks, then call them statistically significant because of their low base probability of appearance and the fact that they deviated at all from the expected mean. Skeptic magazine has a good discussion of the PEAR lab inanity, and I believe James Randi's commentary addresses it a few times.
The claim that PEAR's research wouldn't be reviewed is probably false, by the way. It's most likely that the papers were rejected from mainstream journals for the very reasons I mentioned earlier, or because the PEAR lab had no theoretical explanation for the "results" they observed. Or, of course, it's because their papers seem rather dubious in their lack of data and explanations of how they've arrived at their stated probability values (which I say from having the experience of reading one in a, how shall we say, less than top tier journal). Additionally, the lab's been extremely difficult with regards to their raw data. Randi, for example, has never been able to get ahold of it.
As a graduate student who's been published, I can say this:
We don't get paid jack when we get published.
In fact, for many journals, we have to pay them a substantial per-page fee if we include graphics beyond a fixed (low) limit, or exceed the page count. Even for peer-reviewed books, we don't receive payment for the chapters we contribute. Also, we have to pay for reprints of the article. Oh, and we also have to sign copyright assignment forms that transfer the copyright on our work to the publisher.
Because of this, many people in my department (experimental psychology) have started turning our manuscripts into PDFs and posting them on the web once they're accepted for publication. In that way, we preserve the peer review system and the journal system while simultaneously giving access to those who can't readily acquire the journal. Yes, we know it's not legal, seeing as we signed over all the rights to the publisher, but we feel it's morally appropriate to let scientific knowledge be free.
Indeed. The problem with this project is that the countries where computers are very common (say, the United States) already have effective tsunami systems. The IOC offers tsunami warnings as well in most of the Pacific, and are extending their coverage. The problem is that the third world governments often have problems with disseminating the information, and it's these very same countries that also don't have a lot of modern computers with motion detectors. The places that have effective end-to-end tsunami warning systems don't need this software, and those that don't won't have the hardware necessary to use it.
There's also the concern about sensitivity, of course, including false alarms by people moving their laptops, bumping desks, walking around... some serious testing would be needed to ascertain both the precision (reliability) and accuracy of this system before believing what it says.
I'm guessing you're using phpBB. I've actually been hit by these guys on my boards; it wasn't a problem for me until they started to post. It appears to be actual people and not robots. I should also note I didn't have this problem until I added Google AdSense to my boards. After I did that, I started to get two or three of these spammers each week. Another phpBB board I administer hasn't gotten a spam user yet.
What worked for me was checking the registration e-mail addresses of these people and putting in bans for "*@mail.ru" and "*@*.info". On phpBB, you'll have to manually add these to your ban list table in the forum database. Given that a US board isn't likely to have legitimate users coming from Russia or with.info e-mail addresses (.info generally being the Internet equivalent of the sleazy parts of a big city), I don't think I'm really affecting potential new users. I haven't gotten any complaints or new spam users yet, so my technique seems to be working.
Ah, but MySQL does have graphical administrationtools. They don't come in the server installation package, but they are available and work quite well. They're certainly not the train wreck that SQL Server's tool is.
You're right. In general, MMO companies prefer long-term casual players. Why? They get the same subscription fee from you no matter how much you play. If your account is active and being charged, all the time when you're not actually playing is time when they don't need to be providing bandwidth and support for you. In other words, it's free money for them. Paradoxically, the optimal MMO player (from a profit perspective) is one who keeps an active account but rarely if ever plays. Smedley admitted this in an interview ages ago.
The way to design for this is to make it so that play is short, amusing, and solo friendly, but rapidly gets repetitive. City of Heroes, often attacked for having repetitive content, fits this bill exactly. Such games can be played both by the truly casual player who simply doesn't have the time to play more, as well as by more "hardcore" players who want to keep a secondary MMO around. In both cases, you have players who pay a subscription fee, but "use" very little of it.
Of course, there's nothing necessarily positive about designing for maximized profitability... it just happens that CoH works out that way.
If your definition of "science" is "performing an experiment with the level of replication, control, and measurement expected in the relevant field," then no, they're not doing science. However, any working scientist will tell you that the sorts of exploratory research and demonstrations that the Mythbusters do are actually done -- toy experiments to detected the presence of an effect with a particular manipulation ("pilot studies") are common in experimental psychology, because the costs of doing a full experiment and finding no effect are rather substantial. Take the ping-pong salvage myth, for instance. You certainly couldn't publish a paper on that; there's not enough control for aspects such as water temperature, salinity, and other factors that are relevant, and so you can't generalize the result. However, the fact that they were able to raise the ship suggests strongly that the effect exists, and if you really wanted to explore it further, you could. (Technically, they could probably have just done some math to show it, but that's not nearly as cool.) I would argue that they are doing science -- just not on the level of peer-reviewed outlets, but that's fine given their objectives.
The other important thing the Mythbusters do is to get people thinking scientifically. If you watch an episode and think of ways to blow holes in their design, or ways it could have been done more generally, congratulations -- you're thinking like a scientist. You don't need years of meticulous training in an ivory tower to learn how to do science, and saying otherwise is contributing to the already substantial image problem researchers have.
Slashdot Mirror
...which is that the authors are basing their conclusions on the premise that physiological responses are purely indicative of specific emotional states. This is a position people like Ekman take as well, and it's easily falsified. Russell reviewed the literature around 2000 or so and found that, in fact, emotional displays are at least somewhat socially motivated and don't always equate to specific emotions in a one-to-one fashion (especially in speech). To put it more simply, you may smile because you're happy, but you might also smile because you're being sarcastic, because you're covering up frustration, etc. However, this study rests on the assumption that people only smile because they're happy -- period. (In all fairness, they're not exactly talking about smiling here, but the principle is the same.)
So, sure, if you're willing to accept that premise, then this study is great. If not, it's just another in a long line of studies that suggest, but do not convincingly prove, what emotions can be generated by particular events. Really, this study just again points out how insanely difficult it is to get to a "ground truth" of what emotions people experience.
And yes, I've read the article -- I happen to have electronic access to the journal.
Luckily, if you happen to have a copy of Office, you can go to Help->About and click on the conveniently located link to read the EULA.
Upon reading the EULA for Office 2003, I can't find the clause you mention. It's a pretty standard EULA and is actually surprisingly generous in its licensing terms, especially with regards to clip art (which is more or less "don't use it to violate any depicted entity's trademark, and if you do, we're not liable"). There's nothing in there that I could find about not using Office to hurt Microsoft. Perhaps it's in the French translation?
Actually, record companies advance the band up front, which isn't quite the same as paying them. The advance is paid out of the royalties of the album sales, so in essence, the record company is just giving them a loan. The record companies try to recoup 100% of all expenses out of royalties; this includes music videos, tours, production costs, many forms of advertising, and more. Even the band's producer is paid by the band out of a 100% recouped advance. About the only things that aren't 100% recoupable are actual pressing costs and distribution. Music publishing, you see, has very little to do with most other forms of publishing.
I recommend Donald Passman's "All You Need To Know About the Music Business" for a good overview of what record companies actually do, what the average royalties per CD sold actually are, and how recoupable advances can drive popular bands into bankruptcy. You'll discover all sorts of fun tidbits, like the 20% breakage fee on royalties (a holdover from the days of vinyl that bands are stuck with now). No, I'm not shilling -- I've read it, and it's quite enlightening.
To get back to the overriding question, the answer is almost assuredly yes. Radiohead most likely made more money off the download sales than they would have off a physical CD sale, since their royalties per CD are likely less than $3.
This is a common belief, but it's not true.
The financial aid departments at all universities always engage in specialized calculations for financial aid. While they may offer truly impressive packages to smart people, there's a point at which they're going to insist that someone other than them and the federal government pay. That someone is going to either be you or your parents, and if the coursework is difficult enough that a job is out of the question, and your parents won't front the cash... you're out of luck. My problem wasn't being accepted; my problem was paying for it. This is far more common than you seem to believe. I speak from not only personal experience, but also for several people I know who were accepted to places like Harvard and Stanford with scholarships and federal aid, but couldn't make up the difference.
I'll also point out that, as someone who does hire undergrads for research on grants, we *always* make them work for credit for a semester or two first before putting up some money, and we pay rather poorly ($6.50 an hour, generally, max of 20 hours per week, no benefits). Try making up a few thousand a year in tuition and pay for books and pay for food on that.
The problem is not so much that people are doing research in this field -- people still do research into parapsychology and memetics, for example. The problem is asserting that your theoretical framework is true and correct in the face of serious competition and disconfirmatory evidence. Homeopathy's principle claims are not supported by evidence. As a theoretical framework, it doesn't buy us anything in terms of explanatory power over its primary competitor, the placebo effect. The placebo effect is even more predictive, because it can explain results such as "red and purple liquids, colored by a biologically non-reactive dye, have greater treatment effects than clear ones." How does homeopathy address that? Even clinically, homeopathy fails; its results are on par with what you'd predict from placebo.
I don't mind if people spend time looking for results they may never find. It's true that they might stumble upon something, though the evidence so far suggests that they most likely won't. Given the results thus far, we should definitely consider research into homeopathy very risky, and be mindful of spending money on it. That's an issue of efficient resource allocation, however.
My major problems with researchers into homeopathy is that they often violate the epistemological underpinnings and conventions of science (no special pleading, peer review of results, full disclosure of methods, falsifiable theories and hypotheses, etc.), and that they often make assertions that go far beyond, or run completely counter to, the results of their studies. Those two problems cut to the core of why it's a pseudoscience: it claims to be a science, and sometimes even puts on the airs and trappings of scientific pursuits, but it doesn't follow the same epistemological rules and therefore is *not* science.
The Giant Phone Bills of Doom seem to be iPhone specific. I have an unlimited data plan from AT&T, I use it with my HTC Hermes, and my phone bill has never exceeded 25 pages. And that's with three different phone lines on the bill. :)
And yes, before you ask, the AT&T billing system does do detailed data billing for me. Perhaps it's something different in the way the iPhone interfaces with the GSM network, or perhaps iPhone data plan users get different billing procedures.
I can definitively say, however, that any smartphone should have its data connection turned off overseas, or especially on a cruise. Roaming rates on cruise ships are well and truly insane. Roaming data rates in general are disturbingly high.
You can still get time and temperature (preceded by an advertisement for Captain D's) at:
901-526-5261
It's commonly known around Memphis, TN -- at least among those who know about it -- as "JAMJAM1".
Ekman's research is based primarily on the theoretical principle that facial emotional displays are an automatic, uncontrolled process -- in other words, that you smile reflexively because you're happy. In this paradigm, attempting to restrict the display of a facial display will produce a strange expression that'll be easily recognized as fake.
Obviously, that doesn't explain everything. Much to my chagrin, I actually do research in this messy field, and what we've found is that the Ekman approach isn't as good as one would think. One of the issues of Ekman's research is that it's typically done on staged photographs. When we use emotion raters trained on Ekman's Facial Action Coding System to judge the emotional displays in videos of real people making real displays, the inter-rater reliability falls through the floor. (It's even worse if you split off from Ekman's "basic emotion" categories, which are of dubious utility in the real world anyway.) It's not as bad as untrained raters, but it's still not great. This evidence suggests that there's something else going on besides an automatized process. Russell, another researcher in this field, purports an alternative explanation: that emotional displays are in many cases controlled social processes, and can't really be interpreted outside of a social context.
In any event, the parent is right. The claims on "micro-expressions" and Ekman's FACS in general are not nearly as defensible as their proponents in the TSA would like you to believe. From a signal detection standpoint, the problem isn't so much that you'll have misses, but that you'll have false alarms. More worrying, though, is our line of research shows that the miss and false alarm rates are in many ways a function of individual differences. In other words, some raters err more towards too many identifications (high FA rate), and others err towards too few (high miss rate).
As boring as it sounds, more research is needed before this is implemented -- "this" being any security measure based on Ekman's research.
MFC prevents this (though, interestingly, the .NET Framework doesn't :) ) because, internally, the GDI window code isn't thread-safe. If you call windowing functions for a window handle from a different thread than the one that owns that handle, Bad Things Happen. It's a fundamental rule of Windows GUI programming, albeit a somewhat obscure one for most users. Windows XP and Vista are substantially better at dealing with violations of the rule, but Windows 9x OSes would often go down completely if you broke that rule. Apps under XP sometimes work properly, and sometimes blow up with inscrutable exceptions.
.NET programmers should all know, is to marshal the calls onto the appropriate thread for the control. In .NET, that means checking Control.InvokeRequired and using Control.Invoke with an appropriate delegate for the method you want to call if that property is true (i.e., you're not on the thread that created the control). I've not done very much MFC programming at all, so I'm not exactly sure how you'd handle this issue over there -- perhaps with a custom event queue, maybe?
The solution, as
Okay, people keep saying this ... but when I report my "drug sales" net income (after amortizing my .45 and deducting bribes), won't they just turn that right around and charge you with a crime, implicitly requiring you to waive the fifth?
They can't directly charge you with a crime AFAIK, though perhaps they might be able to if you did something really boneheaded, like write "illegal drug sales" in the "type of business" section of your tax return.
However, your tax returns can be used as evidence against you in a criminal case. If the police suspect you might be selling drugs, they can acquire your tax records. If those tax records show exorbitant income for your professed job, that could be used as evidence sufficient to acquire a search warrant, and then you're boned. Certain types of transactions have to be reported and made available to law enforcement agencies as well. The question for the criminal then becomes a risk analysis: is it safer to conceal the income from the IRS, or report it and hope it's sufficiently hidden or justifiable?
This page describes some of the IRS' activities in narcotics enforcement.
Just to clarify, AT&T has basically four different data plans based on the type and -- so they say -- intended usage of the device. The cheapest one is for "smartphones," which covers all phones without keyboards. The "PDA" plan is the next most expensive, for all phones with keyboards (HTC phones, Treos, etc.) except for Blackberries, which have their own plan. The most expensive is the "computer" plan, which covers their HSPDA/EDGE PC Cards for use in laptops. Some of the plans have a tethered option, where if you pay extra, you can use your phone as a network gateway for your computer. I had heard they were planning on dispensing with the Blackberry plan and rolling it into the PDA plan, but that's second-hand from a CSR, so take it with a grain of salt.
The price of the plan is primarily based on the expected data usage. Additionally, all smartphones have to go through a WAP gateway (MediaNET) for their Internet access. It's an open secret that AT&T will hit you up with per-kb data charges or demand you upgrade to a more costly plan if you use a large amount of transfer for your type of device or if they catch you using their direct connection servers on the cheap smartphone data plan.
Unless Apple has a special deal for the iPhones, they'll be charged either per-kb or on the PDA Connect plan, which is currently $39.95/mo. for unlimited data. They *might* end up creating an iPhone specific plan on the grounds that they may not support XpressMail/GoodMail/DirectPush and thus will use less data than your average business smartphone, but I doubt it. I don't suspect AT&T will apply an additional fee to iPhone users.
The Win32 API can technically support 2^16 mouse buttons (or 2^32 on Win64) in addition to left, right, and middle. The limit is in the design of the WM_XBUTTON messages, where the high order word of the w message parameter specifies which extra button was clicked. Right now, the OS only generates events for XBUTTON1 and XBUTTON2, the two extra buttons on the Intellimouse and its clones. WM_APPCOMMAND messages are used to handle any excess buttons, which are mapped to specific application commands.
System.Windows.Forms in the Framework supports all 5 mouse buttons in its various Mouse events.
Every MMO I've played has had very strict internal oversight on its GMs and devs. They can't use their GM account to play non-GM characters. They can't play with their normal player account on servers that they have GM duties on (I'm aware that wouldn't work with EVE, though). When a "canon character" is needed for an event, there's a sign-out and approval process, and the people who are allowed access to such characters are limited. Log audits of GM abilities are performed periodically to make sure that people aren't abusing their powers.
Yes, certainly, some things slip through -- on Firiona Vie on EQ, someone I played with regularly was given a 100% CSR rez item by a guide, with the understanding that it'd only be used to rez people who died due to bugs. That said, strong internal controls would help protect everyone against the sort of malfeasance that may, or may not, be happening on EVE. This election likely won't solve anything, especially since the winners will likely be members of the large corporations/alliances, who are the people being accused of cheating in the first place.
The fact that the APIs are published is immaterial, really. I know of at least one company's software that has a similar clause in its EULA to the Visual Studio one, and they don't do even the slightest protection of their fully documented APIs. Nevertheless, using them would violate the EULA if I wasn't licensed to do so. That's the contract I entered with that company. If I want to use those APIs, I need to pay them more money for the license to do so. This isn't anything terribly new, and even "good guys" do it now and then to protect a revenue stream (cf. MySQL's multiple licenses).
;)
For those not clear on the situation, the short of it is this. TestDriven.NET is an add-in for Visual Studio. Visual Studio Express has a "technical limitation" that ostensibly prevents the loading of add-ins (removal of the Add-In Manager, I believe). The EULA states that:
"...you may use the software only as expressly permitted in this agreement. In doing so you must comply with any technical limitations in the software that only allow you to use it in certain ways... You may not work around any technical limitations in the software."
Constructing an add-in that can be loaded by Express is presumably a violation of the EULA for Express, because you're working around the technical limitation (weak though it may be) in the software that blocks the loading of add-ins. Technically speaking, anyone who uses it with Express is also violating the EULA. The best argument, IMO (and IANAL), is going to be that disabling the Add-In Manager isn't really a technical limitation against the loading of add-ins, since they can be loaded programmatically. It's a technical limitation against end-users manually loading add-ins.
I regularly read and edit Word and Excel documents on my Cingular 8125. It's often much easier for me to make modifications to an e-mailed document or write a new document on it than to pull out my laptop and boot it up -- seeing as my phone is always with me, while my laptop isn't. As a graduate student, I also find it's handy for taking notes in class or in meetings so I can e-mail them later or sync them up to my computer for further editing and printing. I won't say that this is true for everyone, but I will say that I feel my productivity has gone up since getting my 8125 (to replace my Treo 600, which itself was a big leap over my old Nokia, uh, "dumbphone").
That said, I think the iPhone will fail as a corporate device not for two reasons, neither one of which is not having Office. One, it's unlikely it will support direct push e-mail on Exchange without GoodLink or some custom carrier solution (such as Cingular XpressMail), which will increase internal support costs by interfering with platform standardization. This is basically the same reason why other software and hardware monocultures appear in corporations. Second, the software lockdown aspect, while potentially able to be circumvented, will interfere with custom corporate applications, which are a common thing to put on corporate-owned phones. If nothing else, the fact that the phone runs on OS X will require new versions of these custom applications, which gets back to increasing costs.
The iPhone's neat, sure. For people out of the business environment and who have a whole lot of money to spend, I'm sure it'll be wonderful. On the other hand, businesses will keep buying the same easy to integrate devices they've been buying unless Apple can give them an extremely compelling reason to switch. So far, they haven't done that.
The methodology wasn't flawed, so much as the analysis and the conclusions drawn from it.
A PEAR experiment involved a participant attempting to influence a random number generator (essentially) in a pre-specified direction over a large number of trials. Because random events are, by nature, random, you can get streaks that are above or below the mean. If you analyze a large enough sample, these streaks can become statistically significant, even though they're essentially meaningless and practically insignificant -- it's similar to the fact that any deviation from the mean, no matter how small, is statistically significant if you measure the entire population. Additionally, while the probability of any particular streak is low (.5^n is the probability of any number of heads flipped in a row, which gets very small when you talk about enough of them), if you have enough random events, those streaks are pretty much guaranteed to appear.
So, that's the logic of the PEAR data analysis. Collect a huge corpus of random events, look for streaks, then call them statistically significant because of their low base probability of appearance and the fact that they deviated at all from the expected mean. Skeptic magazine has a good discussion of the PEAR lab inanity, and I believe James Randi's commentary addresses it a few times.
The claim that PEAR's research wouldn't be reviewed is probably false, by the way. It's most likely that the papers were rejected from mainstream journals for the very reasons I mentioned earlier, or because the PEAR lab had no theoretical explanation for the "results" they observed. Or, of course, it's because their papers seem rather dubious in their lack of data and explanations of how they've arrived at their stated probability values (which I say from having the experience of reading one in a, how shall we say, less than top tier journal). Additionally, the lab's been extremely difficult with regards to their raw data. Randi, for example, has never been able to get ahold of it.
As a graduate student who's been published, I can say this:
We don't get paid jack when we get published.
In fact, for many journals, we have to pay them a substantial per-page fee if we include graphics beyond a fixed (low) limit, or exceed the page count. Even for peer-reviewed books, we don't receive payment for the chapters we contribute. Also, we have to pay for reprints of the article. Oh, and we also have to sign copyright assignment forms that transfer the copyright on our work to the publisher.
Because of this, many people in my department (experimental psychology) have started turning our manuscripts into PDFs and posting them on the web once they're accepted for publication. In that way, we preserve the peer review system and the journal system while simultaneously giving access to those who can't readily acquire the journal. Yes, we know it's not legal, seeing as we signed over all the rights to the publisher, but we feel it's morally appropriate to let scientific knowledge be free.
Indeed. The problem with this project is that the countries where computers are very common (say, the United States) already have effective tsunami systems. The IOC offers tsunami warnings as well in most of the Pacific, and are extending their coverage. The problem is that the third world governments often have problems with disseminating the information, and it's these very same countries that also don't have a lot of modern computers with motion detectors. The places that have effective end-to-end tsunami warning systems don't need this software, and those that don't won't have the hardware necessary to use it.
There's also the concern about sensitivity, of course, including false alarms by people moving their laptops, bumping desks, walking around... some serious testing would be needed to ascertain both the precision (reliability) and accuracy of this system before believing what it says.
I'm guessing you're using phpBB. I've actually been hit by these guys on my boards; it wasn't a problem for me until they started to post. It appears to be actual people and not robots. I should also note I didn't have this problem until I added Google AdSense to my boards. After I did that, I started to get two or three of these spammers each week. Another phpBB board I administer hasn't gotten a spam user yet.
What worked for me was checking the registration e-mail addresses of these people and putting in bans for "*@mail.ru" and "*@*.info". On phpBB, you'll have to manually add these to your ban list table in the forum database. Given that a US board isn't likely to have legitimate users coming from Russia or with .info e-mail addresses (.info generally being the Internet equivalent of the sleazy parts of a big city), I don't think I'm really affecting potential new users. I haven't gotten any complaints or new spam users yet, so my technique seems to be working.
Ah, but MySQL does have graphical administration tools. They don't come in the server installation package, but they are available and work quite well. They're certainly not the train wreck that SQL Server's tool is.
You're right. In general, MMO companies prefer long-term casual players. Why? They get the same subscription fee from you no matter how much you play. If your account is active and being charged, all the time when you're not actually playing is time when they don't need to be providing bandwidth and support for you. In other words, it's free money for them. Paradoxically, the optimal MMO player (from a profit perspective) is one who keeps an active account but rarely if ever plays. Smedley admitted this in an interview ages ago.
The way to design for this is to make it so that play is short, amusing, and solo friendly, but rapidly gets repetitive. City of Heroes, often attacked for having repetitive content, fits this bill exactly. Such games can be played both by the truly casual player who simply doesn't have the time to play more, as well as by more "hardcore" players who want to keep a secondary MMO around. In both cases, you have players who pay a subscription fee, but "use" very little of it.
Of course, there's nothing necessarily positive about designing for maximized profitability... it just happens that CoH works out that way.
http://anonetnfo.brinkster.net -- The internet the way it was meant to be. Check it out, you may be surprised.
403 Forbidden?
If that's the Internet was meant to be, sign me up!
We seem to be scaling rather nicely.
It's an excellent scalability mechanism -- everyone gets turned away. Bravo, sir.
(I'm sure there's just something wrong with the server right now, but between the topic and this post, I couldn't resist some gentle joshing.)
If your definition of "science" is "performing an experiment with the level of replication, control, and measurement expected in the relevant field," then no, they're not doing science. However, any working scientist will tell you that the sorts of exploratory research and demonstrations that the Mythbusters do are actually done -- toy experiments to detected the presence of an effect with a particular manipulation ("pilot studies") are common in experimental psychology, because the costs of doing a full experiment and finding no effect are rather substantial. Take the ping-pong salvage myth, for instance. You certainly couldn't publish a paper on that; there's not enough control for aspects such as water temperature, salinity, and other factors that are relevant, and so you can't generalize the result. However, the fact that they were able to raise the ship suggests strongly that the effect exists, and if you really wanted to explore it further, you could. (Technically, they could probably have just done some math to show it, but that's not nearly as cool.) I would argue that they are doing science -- just not on the level of peer-reviewed outlets, but that's fine given their objectives.
The other important thing the Mythbusters do is to get people thinking scientifically. If you watch an episode and think of ways to blow holes in their design, or ways it could have been done more generally, congratulations -- you're thinking like a scientist. You don't need years of meticulous training in an ivory tower to learn how to do science, and saying otherwise is contributing to the already substantial image problem researchers have.