But as an above post said. A person can stop being a layman. But no matter how much they study that can't see a Jebus.
Not so sure about that one. If you look at it from a strictly scientific point of view, you may well end up concluding that a person who dedicates themselves to religion hard enough may well end up having an experience that they can only explain religiously. There's plenty of strange psychological states that you can put yourself into if you go about it in the right way. Speaking in tongues, spirit possession, mysticism—there are plenty of people who will participate in these and take it as evidence of something akin to "seeing a Jebus."
Yeah, I know you and I can propose scientific explanations for what they experience; but do you seriously think it's unreasonable for a rural Haitian with no scientific education to accept the local explanation that those guys acting funny in the Vodou ceremonies were being possessed by spirits? What would you do if you had the same education and life experiences?
(And for the record, scientific understand of spirit possession is fairly spotty. The general shape of the theory I recall is that practitioners go into mental states that are similar to psychosis in Western societies, but in a culturally appropriate fashion according to local norms. I.e., from the point of somebody like you or me, these people have a mental switch that the priest can flip to turn the crazy on and off. There's also dudes who just fake it, however.)
Ken Thompson was once asked what he would do differently if he were redesigning the UNIX system. His reply: "I'd spell creat with an e."
--Kenneth Thompson
Apple has some work on this area that is instructice to consider. Have a look at Applescript, which is an (admittedly crappy) simple language for scripting GUI applications in a standard manner. I do hate Applescript very much, and consider it one of the worst languages ever, but still, the basic mode is the important thing here: GUI applications expose their functionality in a standard fashion through a textual interface that supports programming.
It has to do with JavaScript because it lets you write code that is very, very similar in terms of how it looks (var i = 1; or new {.member1 = "sfdkl",.member2 = "asfdsfdklasfd" };) and a little in the way it acts. No, its not dynamically typed so I can't use i from above and assign a string to it. But I'm talking about syntax here, baby.
So you basically are bitching about things that don't matter, like whether you have to use the keyword "var" in an implicitly typed variable declaration, even when the actual type systems work nothing alike.
Now, I fail to see how anonymous types really help out in writing database-oriented applications. Presumably those types are going to be created through Emit, in which case you can give them a name (and even if you don't this isn't so much a problem, the problem is when people just shit out anonymous types on the fly in code), or you're writing them in the code and in that case you can jsut create a quick struct people can reference.
Anonymous types, from what it looks, simply ensure that you automatically have one-off types that are optimal for ad-hoc queries.
See it this way: if you have a table with n columns, there are n^2 possible subsets of those columns that could be used in a SQL query. Creating structs manually to fit the results of queries is only practicable if your application is only going to be using a few of those subsets, over and over. And it gets even worse when you consider that the SELECT clause of a statement can use scalar operations to expand on the number of columns; basically, SQL naturally generates an infinite set of types.
So when the range of column subsets that your application uses is larger, less predictable and often one-off, you're better off with the anonymous types. A typical example would be something like a ROLAP application, which routinely runs ad-hoc queries against arbitrary column subsets of large tables with lots of columns and rows. You could easily have hundreds of different queries that use different subsets of the columns of a star schema, so that no one combination occurred more than a handful of times.
I know that I can tell you that in Java, one thing people often do in this ad-hoc query situation is to represent database query result rows either as an array of the column values, a list of the column values, or a map from the column name or descriptor to the value. Anonymous types sure look a lot better for this.
If you fail to see how the anonymous types and implicit typing is really close to JavaScript in syntax and in functionality then I don't know what to tell you.
One of them is statically typed, the other is dynamically typed; how the heck is that "close in functionality"?
And again, who a damn about the syntax? The crucial thing is that the.NET version, thanks to static typing, will immediately reject incorrect programs that Javascript just cannot.
Here is what I hate about C#, and I think it is JavaScript infection (they're doing the same to VB, which is pretty much now C# without braces): implicit and anonymous types. That shit is gross.
Ok, so I went and looked up implicit types in.NET, and it turns out that they're nothing like Javascript. Javascript is a dynamically typed language;.NET languages are statically typed, but the compiler can infer the correct type of the variables.
I also went and looked up anonymous types, and they clearly seem to help in writing database-oriented applications. Object-oriented code that's written to use an object-relation mapper very often suffers from the defect that it has to pull all of the columns of a table to construct the objects, even when the caller may only need a subset of those columns. By writing the clients so that their type specifies only the table attributes that they actually need, that allows for performance optimizations.
The.NET implementation doesn't look like it goes all the way in this regard, but hey, they're trying—something that can't be said for any other mainstream language with their crappy "SQL is just strings and prepared statements" nonmodel.
So again, WTF does any of this have to do with Javascript?
GiveWell is the strictest website I've found about analyzing charities for cost-effectiveness. I'm not vouching for their accuracy, but go have a look around their website—they try to be pretty open about how they judge things.
Would you rather lose a few years off your old age (after having retired from a career and seen and played with your grandkids), or face a very small chance of dying soon after a devastating nuclear disaster? Joe Schmoe has chosen the former, and it's entirely logical. Known risk vs. small chance of catastrophe.
I'd rate you up if I could. I disagree that that way of thinking is logical - it's the same irrational reasoning that makes people buy lottery tickets. A known small loss vs. a minuscule chance of a big payoff. But yes, that is the way most people think.
The same? It's completely different. What GP describes is risk aversion. These people are accepting an outcome lower than the claimed expected value of nuclear power because of the uncertain possibility of a terribly adverse outcome. And let's stress the uncertainty part here: it's easy to come up with much different risk/benefit analyses for this stuff.
The lottery buyers, in contrast, are going for a small-but-nearly-certain loss because of the remote chance of a big payoff. That's completely different!
No. The Red Cross actually suggests that you donate often to keep their accounts ready to deal with a disaster at a moments notice instead of them having to wait for donations. So it's possible that money they received because of a flood of donations after, for instance, the earthquake in Haiti is being funneled into relief destined for Japan right now.
Money is fungible, which people all too often fail to understand, and they talk about what particular money was used for what.
Anyway, this leads naturally to two other important pieces of advice if you choose to donate now:
Don't donate consumption goods like canned food or old clothes; donate money, or liquid assets like stock or bonds.
Don't earmark your donation to help for one particular disaster, or one particular type of help.
Red Cross and other charities will have a better idea of what goods are needed than you will, and they can get much better prices for them than you can. Every time a disaster comes around, charities are flooded with people's castaway goods that they have to spend time sorting through to find out what's actually useful, and too often they get too much of some things, and not enough of others.
But when you look at the other planets and how they'd have to go around in looping circles that run along a spiral across the sky, you'll soon notice that it's much more "elegant" to just move the sun into the center. Now they all circle around in (more or less) circular motions.
Except that the Copernican model isn't simpler than Ptolemaic geocentrism. Look it up.
There's also the whole fact that heliocentrism and geodynamicism contradict Aristotelian mechanics. We have Newtonian mechanics today, but well, it had to be invented. And what about that pesky stellar parallax that heliocentrism predicts, but which was never observed until the 1800's?
And it's funny that you talk about "circles" here, because one of the things that really killed geocentrism was Kepler's demonstration that the trajectories were elliptical, not circular. That's one of the big things that messed up the geocentric models—the heliocentric models were better able to cope with that.
It's hardly "a huge amount of evidence" though. It could also mean that the earth is going around the sun, or the two are stationary and the earth is simply rotating on the spot, or the two are revolving around each other while both spin (which is actually our closest understanding of the truth I suppose), etc.
All of these require you to explain how come when you drop a rock from a tower it falls straight down to the feet of the tower, and not away from it; the third one, in addition, is considerably more complex than the first two. The ancient Greeks considered the question of whether the Earth moves in substantial detail, and the general conclusion was that the Earth doesn't move.
Galileo needed to introduce concepts of inertia and relativity of motion as ad hoc hypotheses to work around this objection, but not until Newton was there a good theory of this. Galileo used the example of a man dropping a coin while standing on a moving boat to discredit the tower argument, but well, he was basically claiming that the Earth and everything in it was always in perpetual, undetectable motion, and had no good argument for that other than the fact that it was required to make his heliocentric model work—which had other problems: it predicted stellar parallax, which nobody at the time could observe no matter how hard they tried. (And as a side note, Galileo's independent argument for the movement of the Earth was that it caused tides—an argument which was later shown to be wrong!)
Also, it's important to keep in mind how the Aristotelian theory of mechanics works. Basically, cast in modern terms, it assumes that there is a privileged frame of reference that is at true rest, and that it coincides with the Earth. You don't overturn Aristotelian mechanics in one decade.
And before anybody tries, don't knock the ancient Greeks. Think, for example, about one of Aristotle's arguments on why the Earth is spherical: because the shadow on the moon during a lunar eclipse is round, independently of where on Earth the eclipse is observed from. This explanation presupposes the same theory that we hold to this day about lunar eclipses: that the Earth casts a shadow, and that that shadow sometimes falls on the moon. These dudes supported their theories with careful argument, and figured out tons of stuff that we still believe.
The sandbox issue is what makes it a pain in the behind, but trying to do professional photo work on a iPad is a joke, since the device itself is little more than a underpowered stripped down PC with a ARM processor.
Have you ever used Lightroom? There's this part of Lightroom called the Library module, which has functions like rating (1-5 stars), tagging, metadata management (e.g., annotate photos with titles, locations, descriptions, etc.), and a few simple global image adjustments like preset white balance (no option to set a custom one) and brighter/darker exposure adjustment. Lightroom being a non-destructive photographic workflow tool, these changes don't get done to the original files; they get stored to Lightroom's internal database.
That stuff could easily run on an iPad, so you could review, rate, tag and annotate your images on the tablet, then import your work on the field. You need the powerful machine to produce the final output, but the no less important process of figuring out which of the thousand images you shot today are "keepers," and tagging your collection so that you can later find photos when you need them, that can be done on the tablet.
Apple, has NEVER been bleeding edge, just like Nintendo. They always pick the parts and technology that they can make a profit on. You can see this with the upgrade cycle Apple and Nintendo uses.
If you define the "bleeding edge" as having the best parts, without consideration about who has the best whole, which, well, misses the forest for the trees. The best computer is not necessarily the fastest one.
Many photographers would like to use an app related to Adobe's Lightroom for immediate importing, tagging, and initial review and screening of photos from a photoshoot. But because of Apple's restrictions on direct writing of files and a few other technical issues, this is not apparently possible at this time, according to Adobe.
Do you have a link for that? Because all of that is possible currently. There is no restriction on "direct writing of files".
You do currently have to use the iPad photo app to read images from the Apple SD reader. But there's a third party SD reader which works, that someone like Adobe could leverage into a full-blown photo management/review application.
I was curious, so I Googled this up, and got this open letter from Camera Bits, a maker of photographic workflow software. I can't vouch whether the information there is still accurate, but the claim is that iPad applications don't have direct access to the camera's memory card's filesystem, so photos must first be imported through Apple's Photo Library application; and then, the only way for a third-party app to access those photos is to use an Apple-provided API and UI widget that is not adequate for advanced users.
Even if I had the other whizbangs for editing, cataloging, tagging, etc, I wouldn't use an iPad because I don't know that the image I see will match print. There's no way to calibrate it. Even software calibration would be a slight improvement, but without hardware calibration it's a bust for me and should be a bust for any photographer who takes his or her work seriously.
What exactly are you trying to do that requires the iPad color to be accurate on-site? I can see two types of use:
Doing non-color critical work on the field (e.g., review tag, rate, rotate, crop). It would be great to have the ability to do this sort of thing on the iPad before you get to your real computer, and then import your work from iPad to Lightroom.
Showing photos to clients. In this case you can't guarantee color accuracy anyway because you may not be able to control ambient lighting, so the bar there is much lower: the color has to be "pretty," not 100% accurate.
The N is not a number, it stands for "non-deterministic", meaning that problems in NP can by solved by non-deterministic turing machine (which we don't know how to build) in polynomial time (the "P" stands for polynomial). Problems in P can be solved in polynomial time by a regular (deterministic) turing machine.
And the simpler way of explaining this (not guaranteed to be 100% accurate) is the following: a non-deterministic Turing Machine is basically an infinitely parallel computer, in the sense that it can spawn any finite number of truly parallel threads at no cost. An NP problem is one that you can solve in polynomial time in such an infinitely parallel machine; the general nondeterministic algorithm is just to spawn one thread for each candidate solution, have each thread simply check whether the candidate passes, and return a successful try.
If P=NP, what that means is that there is an efficient single-threaded simulation of an infinitely parallel computer.
My old CS professor called NP "non-polynomial" time so NP=P meaning something like being able to reduced 2^N computations to N^power computation.
Wow, your CS prof messed up big time in explaining that. The question whether NP is polynomial or not is one of the biggest freaking open questions in CS.
The whole "speedy solution" vs. "speedy check" way of describing P vs. NP is not the one I prefer, because it is a bit derivative. The root of the NP class comes from a concept in automata theory called a Nondeterministic Turing Machine. The simple explanation of this is that it's a mathematical model of an infinitely parallel computer, in the sense that it can perform any finite number of computations in parallel (there's no upper bound on the number of threads, and these threads are truly parallel).
NP is the class of problems that such a computer could solve in polynomial time. Since such a computer can spawn a separate thread for each candidate answer, the alternative statement of this is that it's the problems for which proposed solutions can be checked in deterministic polynomial time—since the generic nondeterministic algorithm for solving any problem is "try all candidate solutions in parallel, and pick one that checks out."
In other words: if P=NP, that means that there is a relatively efficient single-threaded simulation of an infinitely parallel computer.
The definition of rape in Sweden is a lot more liberal than in the UK (or US, for that matter). I think the charge is something along the lines of "having sex without a barrier protection method on the assumption that the parties involved do not sleep around and then subsequently finding out that at least one does sleep around, thus increasing the risk of STDs to the injured party".
It's not, and it's not. Assange has been charged with rape, and the allegations include the use of force to overcome resistance.
Basically, Assange's lawyers have made up a lot of shit about this case, and people like you are eating it up.
Oddly, I wasn't aware that you could be extradited from the UK to face a charge for something that isn't illegal in the UK - I think they're trying (and succeeding, so far) to peg it onto a "sexual assault" charge
That 16-vs.-24 page that you quote is about music recording and production. In that application, yes, 24 bit audio is very valuable. Similar issues apply in digital photography: 16 bits per color channel are valuable for capture and subsequent editing, but for showing the end product, 8-bit JPEGs are fine.
It's not just Javascript. We've been living with abysmally sub-optimal implementations of dynamic languages for about two decades.
There are lots of techniques for implementing dynamic languages efficiently. Lisp implementations from the 80s, in particular, used tons of techniques to give really good performance. However, all of that has been lost from the collective consciousness of programmers. The leading dynamic languages are Javascript, Python, Ruby and Perl, and their leading implementations have always sucked.
But basically, as much as Javascript sucks, since it's the web standard and the web is seriously important (read mega-$$$$), there is serious effort to make it go as fast as possible, so techniques are being rediscovered or found.
Apple computers come with tons of crap I'll never use so they aren't immune either. Examples would be: Garageband, iTunes, iMovie and ten other iNoun applications when I use it for work.
Yeah, but those don't start up whenever you boot up your newly bought computer and hog resources to make it nearly unusable. They're also not trial applications that unpromptedly pester you to buy the full thing (though granted, they sometimes promptedly push other Apple stuff on you, like Quicktime Pro).
You can also completely ignore all of the iApps if you want. You can even trivially delete them—drag them to the trashcan, and presto. Software updates for the iApps also happen through the centralized Apple software update tool, so you don't have half a million per-app software update windows popping up every day.
The bloatware problem isn't preinstalled add-on software. It's intrusive preinstalled add-on software.
Looks like he's trolled about 620 thousand people. I think the best way to socially engineer a geek is to make them feel superior to you, because they cannot resist loudly correcting you and assuming you're an idiot from that point on.
But in this case, it makes sense. "What exactly is a galaxy" doesn't matter to cosmology or astrophysics; it's just a definition, or a convention to call something as "galaxy". What should be defined as "galaxy" doesn't matter, except for consistency reasons; and that is why a vote, or an attempt to reach consensus, is the best way to reach a good definition.
But the question I'm going to raise now is how much this "consistency" that you're talking about really matters. You may say something vague or conventional like "it prevents confusion," but then I'm going to question whether that is in fact really so, illustrated by actual examples of people getting confused.
There's nothing wrong with vague categories defined by central examples. There's usually no need to draw a definite line as long as the clear examples outnumber the gray cases by a healthy margin.
Yep, that's why currently many are being reclassified based on genetic comparison, although I'm not sure what the cutoff is and whether the same one is used by all biologists.
The fundamental idea of evolutionary theory is that any two organisms have a common ancestor. Think about that closely: it entails that there cannot be a principled line to be drawn between species, because no matter how you propose to draw such lines, organisms of any "species" must have bred with members of an ancestor species.
The only strictly factual claims that can be made are about which organisms were the ancestors of which. "Species" are just vague areas of interest in the tree of life; no lines need to be drawn. There's reclassification going on in light of genetics, but the only really meaningful cases are the ones that involve changing hypotheses about the shape of the evolutionary tree: "The most recent common ancestor of A and B is D, not C."
Slashdot Mirror
But as an above post said. A person can stop being a layman. But no matter how much they study that can't see a Jebus.
Not so sure about that one. If you look at it from a strictly scientific point of view, you may well end up concluding that a person who dedicates themselves to religion hard enough may well end up having an experience that they can only explain religiously. There's plenty of strange psychological states that you can put yourself into if you go about it in the right way. Speaking in tongues, spirit possession, mysticism—there are plenty of people who will participate in these and take it as evidence of something akin to "seeing a Jebus."
Yeah, I know you and I can propose scientific explanations for what they experience; but do you seriously think it's unreasonable for a rural Haitian with no scientific education to accept the local explanation that those guys acting funny in the Vodou ceremonies were being possessed by spirits? What would you do if you had the same education and life experiences?
(And for the record, scientific understand of spirit possession is fairly spotty. The general shape of the theory I recall is that practitioners go into mental states that are similar to psychosis in Western societies, but in a culturally appropriate fashion according to local norms. I.e., from the point of somebody like you or me, these people have a mental switch that the priest can flip to turn the crazy on and off. There's also dudes who just fake it, however.)
Ken Thompson was once asked what he would do differently if he were redesigning the UNIX system. His reply: "I'd spell creat with an e." --Kenneth Thompson
"creet"? That's even worse!
Apple has some work on this area that is instructice to consider. Have a look at Applescript, which is an (admittedly crappy) simple language for scripting GUI applications in a standard manner. I do hate Applescript very much, and consider it one of the worst languages ever, but still, the basic mode is the important thing here: GUI applications expose their functionality in a standard fashion through a textual interface that supports programming.
It has to do with JavaScript because it lets you write code that is very, very similar in terms of how it looks (var i = 1; or new { .member1 = "sfdkl", .member2 = "asfdsfdklasfd" };) and a little in the way it acts. No, its not dynamically typed so I can't use i from above and assign a string to it. But I'm talking about syntax here, baby.
So you basically are bitching about things that don't matter, like whether you have to use the keyword "var" in an implicitly typed variable declaration, even when the actual type systems work nothing alike.
Now, I fail to see how anonymous types really help out in writing database-oriented applications. Presumably those types are going to be created through Emit, in which case you can give them a name (and even if you don't this isn't so much a problem, the problem is when people just shit out anonymous types on the fly in code), or you're writing them in the code and in that case you can jsut create a quick struct people can reference.
Anonymous types, from what it looks, simply ensure that you automatically have one-off types that are optimal for ad-hoc queries.
See it this way: if you have a table with n columns, there are n^2 possible subsets of those columns that could be used in a SQL query. Creating structs manually to fit the results of queries is only practicable if your application is only going to be using a few of those subsets, over and over. And it gets even worse when you consider that the SELECT clause of a statement can use scalar operations to expand on the number of columns; basically, SQL naturally generates an infinite set of types.
So when the range of column subsets that your application uses is larger, less predictable and often one-off, you're better off with the anonymous types. A typical example would be something like a ROLAP application, which routinely runs ad-hoc queries against arbitrary column subsets of large tables with lots of columns and rows. You could easily have hundreds of different queries that use different subsets of the columns of a star schema, so that no one combination occurred more than a handful of times.
I know that I can tell you that in Java, one thing people often do in this ad-hoc query situation is to represent database query result rows either as an array of the column values, a list of the column values, or a map from the column name or descriptor to the value. Anonymous types sure look a lot better for this.
If you fail to see how the anonymous types and implicit typing is really close to JavaScript in syntax and in functionality then I don't know what to tell you.
One of them is statically typed, the other is dynamically typed; how the heck is that "close in functionality"?
And again, who a damn about the syntax? The crucial thing is that the .NET version, thanks to static typing, will immediately reject incorrect programs that Javascript just cannot.
Here is what I hate about C#, and I think it is JavaScript infection (they're doing the same to VB, which is pretty much now C# without braces): implicit and anonymous types. That shit is gross.
Ok, so I went and looked up implicit types in .NET, and it turns out that they're nothing like Javascript. Javascript is a dynamically typed language; .NET languages are statically typed, but the compiler can infer the correct type of the variables.
I also went and looked up anonymous types, and they clearly seem to help in writing database-oriented applications. Object-oriented code that's written to use an object-relation mapper very often suffers from the defect that it has to pull all of the columns of a table to construct the objects, even when the caller may only need a subset of those columns. By writing the clients so that their type specifies only the table attributes that they actually need, that allows for performance optimizations.
The .NET implementation doesn't look like it goes all the way in this regard, but hey, they're trying—something that can't be said for any other mainstream language with their crappy "SQL is just strings and prepared statements" nonmodel.
So again, WTF does any of this have to do with Javascript?
Really? You find that more offensive than genocide, pedophilia, and split infinitives put together?
What's so offensive about wanting to brutally genocide pedophiles?
GiveWell is the strictest website I've found about analyzing charities for cost-effectiveness. I'm not vouching for their accuracy, but go have a look around their website—they try to be pretty open about how they judge things.
I bring that up as background to this: GiveWell is recommending that people do not donate money specifically for Japan disaster relief, for reasons they explain in that link. Their recommendation is that if you want to donate to a disaster relief charity at this time, you just make a general donation to Doctors Without Borders.
I should stress: click on that second link and at least skim through it.
Would you rather lose a few years off your old age (after having retired from a career and seen and played with your grandkids), or face a very small chance of dying soon after a devastating nuclear disaster? Joe Schmoe has chosen the former, and it's entirely logical. Known risk vs. small chance of catastrophe.
I'd rate you up if I could. I disagree that that way of thinking is logical - it's the same irrational reasoning that makes people buy lottery tickets. A known small loss vs. a minuscule chance of a big payoff. But yes, that is the way most people think.
The same? It's completely different. What GP describes is risk aversion. These people are accepting an outcome lower than the claimed expected value of nuclear power because of the uncertain possibility of a terribly adverse outcome. And let's stress the uncertainty part here: it's easy to come up with much different risk/benefit analyses for this stuff.
The lottery buyers, in contrast, are going for a small-but-nearly-certain loss because of the remote chance of a big payoff. That's completely different!
No. The Red Cross actually suggests that you donate often to keep their accounts ready to deal with a disaster at a moments notice instead of them having to wait for donations. So it's possible that money they received because of a flood of donations after, for instance, the earthquake in Haiti is being funneled into relief destined for Japan right now.
Money is fungible, which people all too often fail to understand, and they talk about what particular money was used for what.
Anyway, this leads naturally to two other important pieces of advice if you choose to donate now:
Red Cross and other charities will have a better idea of what goods are needed than you will, and they can get much better prices for them than you can. Every time a disaster comes around, charities are flooded with people's castaway goods that they have to spend time sorting through to find out what's actually useful, and too often they get too much of some things, and not enough of others.
But when you look at the other planets and how they'd have to go around in looping circles that run along a spiral across the sky, you'll soon notice that it's much more "elegant" to just move the sun into the center. Now they all circle around in (more or less) circular motions.
Except that the Copernican model isn't simpler than Ptolemaic geocentrism. Look it up.
There's also the whole fact that heliocentrism and geodynamicism contradict Aristotelian mechanics. We have Newtonian mechanics today, but well, it had to be invented. And what about that pesky stellar parallax that heliocentrism predicts, but which was never observed until the 1800's?
And it's funny that you talk about "circles" here, because one of the things that really killed geocentrism was Kepler's demonstration that the trajectories were elliptical, not circular. That's one of the big things that messed up the geocentric models—the heliocentric models were better able to cope with that.
So, it's not that simple, dude.
It's hardly "a huge amount of evidence" though. It could also mean that the earth is going around the sun, or the two are stationary and the earth is simply rotating on the spot, or the two are revolving around each other while both spin (which is actually our closest understanding of the truth I suppose), etc.
All of these require you to explain how come when you drop a rock from a tower it falls straight down to the feet of the tower, and not away from it; the third one, in addition, is considerably more complex than the first two. The ancient Greeks considered the question of whether the Earth moves in substantial detail, and the general conclusion was that the Earth doesn't move.
Galileo needed to introduce concepts of inertia and relativity of motion as ad hoc hypotheses to work around this objection, but not until Newton was there a good theory of this. Galileo used the example of a man dropping a coin while standing on a moving boat to discredit the tower argument, but well, he was basically claiming that the Earth and everything in it was always in perpetual, undetectable motion, and had no good argument for that other than the fact that it was required to make his heliocentric model work—which had other problems: it predicted stellar parallax, which nobody at the time could observe no matter how hard they tried. (And as a side note, Galileo's independent argument for the movement of the Earth was that it caused tides—an argument which was later shown to be wrong!)
Also, it's important to keep in mind how the Aristotelian theory of mechanics works. Basically, cast in modern terms, it assumes that there is a privileged frame of reference that is at true rest, and that it coincides with the Earth. You don't overturn Aristotelian mechanics in one decade.
And before anybody tries, don't knock the ancient Greeks. Think, for example, about one of Aristotle's arguments on why the Earth is spherical: because the shadow on the moon during a lunar eclipse is round, independently of where on Earth the eclipse is observed from. This explanation presupposes the same theory that we hold to this day about lunar eclipses: that the Earth casts a shadow, and that that shadow sometimes falls on the moon. These dudes supported their theories with careful argument, and figured out tons of stuff that we still believe.
The sandbox issue is what makes it a pain in the behind, but trying to do professional photo work on a iPad is a joke, since the device itself is little more than a underpowered stripped down PC with a ARM processor.
Have you ever used Lightroom? There's this part of Lightroom called the Library module, which has functions like rating (1-5 stars), tagging, metadata management (e.g., annotate photos with titles, locations, descriptions, etc.), and a few simple global image adjustments like preset white balance (no option to set a custom one) and brighter/darker exposure adjustment. Lightroom being a non-destructive photographic workflow tool, these changes don't get done to the original files; they get stored to Lightroom's internal database.
That stuff could easily run on an iPad, so you could review, rate, tag and annotate your images on the tablet, then import your work on the field. You need the powerful machine to produce the final output, but the no less important process of figuring out which of the thousand images you shot today are "keepers," and tagging your collection so that you can later find photos when you need them, that can be done on the tablet.
Apple, has NEVER been bleeding edge, just like Nintendo. They always pick the parts and technology that they can make a profit on. You can see this with the upgrade cycle Apple and Nintendo uses.
If you define the "bleeding edge" as having the best parts, without consideration about who has the best whole, which, well, misses the forest for the trees. The best computer is not necessarily the fastest one.
Many photographers would like to use an app related to Adobe's Lightroom for immediate importing, tagging, and initial review and screening of photos from a photoshoot. But because of Apple's restrictions on direct writing of files and a few other technical issues, this is not apparently possible at this time, according to Adobe.
Do you have a link for that? Because all of that is possible currently. There is no restriction on "direct writing of files".
You do currently have to use the iPad photo app to read images from the Apple SD reader. But there's a third party SD reader which works, that someone like Adobe could leverage into a full-blown photo management/review application.
I was curious, so I Googled this up, and got this open letter from Camera Bits, a maker of photographic workflow software. I can't vouch whether the information there is still accurate, but the claim is that iPad applications don't have direct access to the camera's memory card's filesystem, so photos must first be imported through Apple's Photo Library application; and then, the only way for a third-party app to access those photos is to use an Apple-provided API and UI widget that is not adequate for advanced users.
Two words: color calibration
Even if I had the other whizbangs for editing, cataloging, tagging, etc, I wouldn't use an iPad because I don't know that the image I see will match print. There's no way to calibrate it. Even software calibration would be a slight improvement, but without hardware calibration it's a bust for me and should be a bust for any photographer who takes his or her work seriously.
What exactly are you trying to do that requires the iPad color to be accurate on-site? I can see two types of use:
The N is not a number, it stands for "non-deterministic", meaning that problems in NP can by solved by non-deterministic turing machine (which we don't know how to build) in polynomial time (the "P" stands for polynomial). Problems in P can be solved in polynomial time by a regular (deterministic) turing machine.
And the simpler way of explaining this (not guaranteed to be 100% accurate) is the following: a non-deterministic Turing Machine is basically an infinitely parallel computer, in the sense that it can spawn any finite number of truly parallel threads at no cost. An NP problem is one that you can solve in polynomial time in such an infinitely parallel machine; the general nondeterministic algorithm is just to spawn one thread for each candidate solution, have each thread simply check whether the candidate passes, and return a successful try.
If P=NP, what that means is that there is an efficient single-threaded simulation of an infinitely parallel computer.
My old CS professor called NP "non-polynomial" time so NP=P meaning something like being able to reduced 2^N computations to N^power computation.
Wow, your CS prof messed up big time in explaining that. The question whether NP is polynomial or not is one of the biggest freaking open questions in CS.
The whole "speedy solution" vs. "speedy check" way of describing P vs. NP is not the one I prefer, because it is a bit derivative. The root of the NP class comes from a concept in automata theory called a Nondeterministic Turing Machine. The simple explanation of this is that it's a mathematical model of an infinitely parallel computer, in the sense that it can perform any finite number of computations in parallel (there's no upper bound on the number of threads, and these threads are truly parallel).
NP is the class of problems that such a computer could solve in polynomial time. Since such a computer can spawn a separate thread for each candidate answer, the alternative statement of this is that it's the problems for which proposed solutions can be checked in deterministic polynomial time—since the generic nondeterministic algorithm for solving any problem is "try all candidate solutions in parallel, and pick one that checks out."
In other words: if P=NP, that means that there is a relatively efficient single-threaded simulation of an infinitely parallel computer.
I was under the impression that Sweden gave him permission to leave the country.
And you were under that impression because Assange and his lawyers told you that.
The definition of rape in Sweden is a lot more liberal than in the UK (or US, for that matter). I think the charge is something along the lines of "having sex without a barrier protection method on the assumption that the parties involved do not sleep around and then subsequently finding out that at least one does sleep around, thus increasing the risk of STDs to the injured party".
It's not, and it's not. Assange has been charged with rape, and the allegations include the use of force to overcome resistance.
Basically, Assange's lawyers have made up a lot of shit about this case, and people like you are eating it up.
Oddly, I wasn't aware that you could be extradited from the UK to face a charge for something that isn't illegal in the UK - I think they're trying (and succeeding, so far) to peg it onto a "sexual assault" charge
Again, your whole premise here is false.
That 16-vs.-24 page that you quote is about music recording and production. In that application, yes, 24 bit audio is very valuable. Similar issues apply in digital photography: 16 bits per color channel are valuable for capture and subsequent editing, but for showing the end product, 8-bit JPEGs are fine.
It's not just Javascript. We've been living with abysmally sub-optimal implementations of dynamic languages for about two decades.
There are lots of techniques for implementing dynamic languages efficiently. Lisp implementations from the 80s, in particular, used tons of techniques to give really good performance. However, all of that has been lost from the collective consciousness of programmers. The leading dynamic languages are Javascript, Python, Ruby and Perl, and their leading implementations have always sucked.
But basically, as much as Javascript sucks, since it's the web standard and the web is seriously important (read mega-$$$$), there is serious effort to make it go as fast as possible, so techniques are being rediscovered or found.
Apple computers come with tons of crap I'll never use so they aren't immune either. Examples would be: Garageband, iTunes, iMovie and ten other iNoun applications when I use it for work.
Yeah, but those don't start up whenever you boot up your newly bought computer and hog resources to make it nearly unusable. They're also not trial applications that unpromptedly pester you to buy the full thing (though granted, they sometimes promptedly push other Apple stuff on you, like Quicktime Pro).
You can also completely ignore all of the iApps if you want. You can even trivially delete them—drag them to the trashcan, and presto. Software updates for the iApps also happen through the centralized Apple software update tool, so you don't have half a million per-app software update windows popping up every day.
The bloatware problem isn't preinstalled add-on software. It's intrusive preinstalled add-on software.
Looks like he's trolled about 620 thousand people. I think the best way to socially engineer a geek is to make them feel superior to you, because they cannot resist loudly correcting you and assuming you're an idiot from that point on.
Well, yeah...
But in this case, it makes sense. "What exactly is a galaxy" doesn't matter to cosmology or astrophysics; it's just a definition, or a convention to call something as "galaxy". What should be defined as "galaxy" doesn't matter, except for consistency reasons; and that is why a vote, or an attempt to reach consensus, is the best way to reach a good definition.
But the question I'm going to raise now is how much this "consistency" that you're talking about really matters. You may say something vague or conventional like "it prevents confusion," but then I'm going to question whether that is in fact really so, illustrated by actual examples of people getting confused.
There's nothing wrong with vague categories defined by central examples. There's usually no need to draw a definite line as long as the clear examples outnumber the gray cases by a healthy margin.
Since when does a scientific term require a dependent theory?
Um, since about the time science became about creating predictive theories?
Yep, that's why currently many are being reclassified based on genetic comparison, although I'm not sure what the cutoff is and whether the same one is used by all biologists.
The fundamental idea of evolutionary theory is that any two organisms have a common ancestor. Think about that closely: it entails that there cannot be a principled line to be drawn between species, because no matter how you propose to draw such lines, organisms of any "species" must have bred with members of an ancestor species.
The only strictly factual claims that can be made are about which organisms were the ancestors of which. "Species" are just vague areas of interest in the tree of life; no lines need to be drawn. There's reclassification going on in light of genetics, but the only really meaningful cases are the ones that involve changing hypotheses about the shape of the evolutionary tree: "The most recent common ancestor of A and B is D, not C."