So maybe they're lazy and can't be bothered to go back indefinitely every time. Or maybe they're aware that if they had a 100% hit rate, it'd give the game away and they'd lose their research grants.
Yes, maybe they are incompetent, and maybe they are corrupt. Very insightful.
Regardless, it's detail that appears to be missing from their description of their experimental method,
Does it? Have you actually read their material, or just the brief summary linked to by Slashdot?
and it's entire legitimate to poke at it. To fail to do so is to be unscientific.
Kind of like accusing them of incompetence or corruption without backing evidence, huh?
Sorry, I accidentally didn't reply to this part of your inane post:
Did he provide the statistics on why the "most likely number of times for them to have seen spikes that large was about ten but they saw sixteen"?
Yes.
"Ten" because they defined a successful spike as being a spike large enough that, if you look at a random time range the size that they look at, the chance of seeing a spike that large or greater is (by elementary probability calculations) five percent. And five percent of 192 is about ten.
You consider an 8% success rate irrelavant? Do you know of any scientific proposal with a 8% success rate that is considered vaild? Are you really arguing that it is?
What?
The theoretical average rate is 10/192. The actual measured rate is 16/192.
The numbers that those rates work out to be is not important.
What's important is that they're different, and what's more important is the degree to which they are different.
That degree can be quantified: Specifically, the chance of 16 or more successes out of 192, when the chance of one success in one attempt is one out of twenty, and the result of each attempt is independent of each other attempt, is approximately eight percent, at least according to this other slashdotter.
Quite the conicidence isn't it? The chances of both succeding in total and the total faliure rate being the same figure. Did this slashdot poster provide you with the calculation? Did he provide you with the time frame for success? Did he provide the statistics on why the "most likely number of times for them to have seen spikes that large was about ten but they saw sixteen"? Did he give you anything at all?
Yes, he did provide the calculation. No, it wasn't your simple inane "success" divided by "attempts". Yes, the near-similarity between the two is a coincidence. No, you saying witty things like "Quite the coincidence, isn't it?" doesn't change the fact that it's a coincidence.
Do you want me to calculate it for you, in painful detail? Well I'm not going to. But I'll tell you how to do it:
Here's a basic law of probability for you: If you have a test that has an P chance of success in any single attempt, and each attempt does not affect any other attempt, then the chance of getting exactly S successes in A attempts is equal to:
( A! * ( P ^ S ) * ( ( 1 - P ) ^ ( A - S ) ) / ( S! * ( A - S )! )
In this case, A is 192 (they tried 192 times), S is 16 (they succeeded 16 times), and P is 0.05 (the chance of a success in any particular attempt was one in twenty).
So plug those numbers into that formula, and you'll get the probability of seeing exactly 16 successes in 192 attempts. If you don't believe me, look it the fuck up before launching into another diatribe.
But that's not what we're interested in - we're interested in the probability of seeing 16 or more successes in 192 attempts.
So, you'll have to calculate that forumula a bunch of times, plugging in the numbers 16, 17, 18, and so forth, up to 192, for the number of successes. Then add up the results of each of those individual calculations.
I'm not about to do that. And I'm not going to guarantee that the number you get by it will be "eight percent". But I will guarantee two things:
(1) It's the correct formula;
(2) If you do get "eight percent", the fact that 16 / 192 is approximately the same is... wait for it... coincidental.
How far back do they go? Do they set the limit before they start looking? If not, and they're prepared to look back indefinitely, they'll always find a spike!
And yet, they don't always find a spike.
In which case, the rate of 'coincident' spikes will of course be significantly higher than random chance!
This is your "then" to your "if" of "if they always find a spike". Since that "if" turns out to be untrue, your "then" is irrelevant.
No, you are wrong. Instead of the chance of succeeding 16 times out of 192 being 8%, the OVERALL success rate is 8%(16/192 * 100 = 8.3%). That is an incredible Failure rate, almost one that could not be attribited to chance. To calculate the chance of sucess vs. failure, you would have to know the time frame in which they designate success, one figure which they do not seem willing to divulge.
You have no idea what you're talking about.
That 16 out of 192 was 16 times that they saw a spike which was large enough so that it should happen, on average, one time out of every twenty.
That is, out of those 192 times, the most likely number of times for them to have seen spikes that large was about ten.
But they saw sixteen.
The chance of seeing sixteen (or more) can be calculated, and it turns out (at least according to another slashdot poster) to be about eight percent.
That 16/192 is around eight percent is merely coincidence. Your calculation has nothing to do with anything at all.
But here's the problem: Starting from a "major event" and working backwards, you will always find a major spike... so long as you look far enough.
Yes, that's true. However, they obviously don't simply just look and look and look further and further back from the event until they find a spike, as evidenced by the fact that most of the time, they don't find spikes.
Assuming that they look for some particular time range around the event - and I haven't been presented with a reason to believe they do otherwise - then the chance of there being a spike during that range is quantifiable. Their claim is that they have exceeded the quantifiable chance, significantly.
ere, they've got two fuzzy criteria -- What is a "major event"?
That's not really relevant. Assuming that their data is not just a fluke (and not incorrect or doctored), then at the very least, they've shown that they predict spikes at a rate significantly greater than random. They seem to be taking for granted that that is (or would be) because of the "major events", but that conclusion is not strictly within the purview of science, based upon the methodology they're using.
How far in advance?
Again, there's nothing that I've seen that suggests their answer to this question is necessarily "fuzzy". And if it's not fuzzy, then its effect is quantifiable.
And much like the canals of Mars, it seems likely that in the random data they are seeing something they want to see.
I think you're guilty, here, of something that a lot of posters in this thread are guilty of: You're assuming that they're not familiar with statistics.
They claim to have statistically quantified the chance of them successfully predicting spikes at the rate that they have, in fact, successfully predicted spikes. The claim being made is that that's a surprisingly low number.
there was a coincident spike in significantly more cases than would have been predicted by random chance.
You replied:
Errr....Random chance doesn't predict anything.
Fine, fine. Let me be painfully explicit:
Define "success" as them saying "maybe there's a spike now", then checking the box, and finding that there is a spike of a certain statistical significance or higher.
Define "failure" as them saying "maybe there's a spike now", then checking the box, and finding that there is not a spike of that certain statistical significance or higher.
They have tried 192 times, "succeeded" 16 times, and "failed" 176 times.
The theory of probability can be used to predict that, if "success" or "failure" is random, your chance of "succeeding" that many times or more in that many attempts is eight percent (assuming calculations done by another/.er are correct).
So they where looking for September 11th then? Or the tsunami? The article indicated that these events were the boxes major triumphs. If thats the case, why do they need the box at all?
It's not entirely clear to me what you're asking.
(1) They noticed that 9/11 happened.
(2) They thought to themselves, "This is a significantly major event that might cause a spike, if the proposed theory is real".
(3) As they always do when they think such things to themselves, they went back and checked the box.
(4) There was a coincident spike.
(5-8) Repeat (1-4) with "the tsunami" in place of "9/11".
(9-large) Repeat again, many more times, for other events, except in most cases, there was no coincident spike.
Yes, in most cases, there was no coincident spike. However - and this is the heart of their claim - there was a coincident spike in significantly more cases than would have been predicted by random chance.
The glaring error in their methodology is that they are Quantifying the un-quantifable. A significant world event is an opinion. What constitues a significant world event is in the eye of the beholder.
As pointed out several times, they don't look for spikes and then look for corresponding major events.
They look for events that they think might cause a spike (if the proposed effect is real), and then see if a spike occurred coincidentally.
So far, they've found such spikes, in this manner, at a rate significantly higher than random chance. Another poster calculated their chances of prediction at this level or higher at eight percent.
Whether them successfully doing something that has an eight percent chance of happening is something to get worked up about or not is up for debate. But your statement about their "glaring error" has nothing to do with anything.
It took me all of 5 seconds to find this article which pretty much debunks the entire project:
The article you link to debunks nothing.
Basically, the argument of the article is this:
(1) He showed me that, coincidentally with such-and-such an event, the RNGs strayed significantly from 50/50.
(2) I showed him that they also did the day before.
(3) I asked him what happened the day before.
(4) He said he didn't know.
(5) Ergo, what a load.
While that argument is seemingly cogent from a simplistic point of view, a little thought reveals that it's basically irrelevant. Specifically, the argument merely debunks a claim that is not being made in the first place.
These spikes happen all the time. No one denies that. Pointing out that - which is all the supposed "debunking" article does - is neither here nor there.
The claim being made is that the spikes happen more frequently than predicted by random chance when "major" events happen. This supposed "debunking" article does not address that claim at all. At all.
The project started with a scientist asking people to come to his lab and concentrate. That one person allegedly influenced the outcome of the random number generator. This would tend to suggest that your influence on the machine is directly related to your distance from it.
How so?
As far as I know, and as far as the article seems to mention, the test did not investigate any potential link between the distance and the outcome.
Yes, the person was in the same lab as the machine. But the test did not then go on to try with the guy in the same block, the same city, the same continent.
All it did was test "the guy thinking about influencing the machine" versus "the guy not thinking about influencing the machine".
To draw any conclusion, based upon that, about distance being involved is unscientific.
Basically, what they did is the following: They looked at the output of this device for "statistical anomalies", i.e. the.5% of the time it's output was out of the range it was expected to be in 99.5% of the time. Then, they looked for "important events" that occurred around that time. They don't have any way of choosing them other than that they happened when this occurred. Finally, they started looking for events *after* it happened, and found those as well. BOOM, it predicts the future.
That's actually utterly false.
As mentioned elsewhere in this thread - multiple times, in fact - they don't look for deviations then look for news that occurred near the time of the deviation.
They look for news that they think might cause deviations if this effect were real, and then look for deviations at the time of that news.
Seriously, they can't. Randi needs a totally controlled environment. The project uses a random generator and world events, both of which I don't think you can get into a controlled envrionment that would satisfy all.
That's not entirely true. The article mentioned and quoted several independent people - scientists, apparently - who claim that merely asking people to think about affecting the numbers generated by random number generators does, in fact, affect the numbers generated by random number generators.
That can easily be tested in a controlled environment. In fact, it's what these scientists in the article claim to have already done.
Uh, that's accomplished, and then some, by instead getting used to a one-standard world.
I officially proclaim us at (or beyond) the point where we can say "screw people with Netscape 4.0 or IE 3 or whatever".
The existing differences between the rendering on the current versions of the main browsers (and most minor browsers too) are so trivial that a completely standards-compliant page can be made to look good in any of them, even if they might look slightly different in each.
IE misinterprets the box model? So what? With reasonably chosen values, things look fine in each. That fancy bevelled border that you want shows up as a plain old border in IE? Who cares? There are worse things than a plain border. And so forth.
A browser, email client, calender, and office suite do not an operating system make.
That minor point aside, a main problem with the whole XUL concept, in terms of extensibility, is the lack of good namespacing (or similar) capabilities in Javascript. If your extension uses a function called "getData", and mine uses a variable called "getData", guess what, there are problems. And nothing is going to tell you "hey, there's a conflict" - the underlying engine will happily call one of them. Maybe not the one you wanted. Global variables further complicate the issue.
Of course, today, this kind of problem is merely one that makes plugin developers scratch their heads and waste a lot of time, until finally they realize "Oh, goddamn, this programming environment is idiotic", and fix the immediate problem.
But if XUL has grand schemes on everything in the entire computer world, this issue has to be addressed.
Another issue is that the base applications don't always give names to widgets that you may want to modify. As far as I know, there is thereafter no way to safely extend those in a plugin - maybe its parent is named "blah", and you can say "modify blah's fourth widget child", but what if another plugin programmer inserted a widget before blah's (initial) fourth, thus making the one that you want to modify the fifth? Then you're modifying some random widget that you didn't intend to, with unpredictable and unintended consequences.
If you think C/C++ buffer overflows are bad, wait until XUL takes over the world. You'll have malware that hijacks function calls, malware that intentionally causes popular plugins to modify things that they weren't intended to modify, and so forth.
Google cache is not particularly useful as a tool against slashdotting when the main objects of interest are the graphics and video stored on the slashdotted server.
Movie review shows aren't allowed to tape their own clips from the movie. The only clips you see in a movie review show are those provided by the studios to review show at the sole discretion of the studios. The studios don't release the whole film to the production companies who make review shows, only the clips the movie companies want to be seen.
You're assuming we're talking about new movies. The post I responded to did not. He was also talking about ripping from DVDs.
The only reason (new) movie review shows only show the clips provided to them is because they're the only clips they have legal access to. If they had legal access to the rest of the movie - as they do when they are reviewing things that have newly come out on DVD - they could legally pick and choose from any portion.
Slashdot Mirror
So maybe they're lazy and can't be bothered to go back indefinitely every time. Or maybe they're aware that if they had a 100% hit rate, it'd give the game away and they'd lose their research grants.
Yes, maybe they are incompetent, and maybe they are corrupt. Very insightful.
Regardless, it's detail that appears to be missing from their description of their experimental method,
Does it? Have you actually read their material, or just the brief summary linked to by Slashdot?
and it's entire legitimate to poke at it. To fail to do so is to be unscientific.
Kind of like accusing them of incompetence or corruption without backing evidence, huh?
Sorry, I accidentally didn't reply to this part of your inane post:
Did he provide the statistics on why the "most likely number of times for them to have seen spikes that large was about ten but they saw sixteen"?
Yes.
"Ten" because they defined a successful spike as being a spike large enough that, if you look at a random time range the size that they look at, the chance of seeing a spike that large or greater is (by elementary probability calculations) five percent. And five percent of 192 is about ten.
"Sixteen" because that's how many they saw.
Any other questions?
Nothing to do with anything at all?
That's correct.
You consider an 8% success rate irrelavant? Do you know of any scientific proposal with a 8% success rate that is considered vaild? Are you really arguing that it is?
What?
The theoretical average rate is 10/192. The actual measured rate is 16/192.
The numbers that those rates work out to be is not important.
What's important is that they're different, and what's more important is the degree to which they are different.
That degree can be quantified: Specifically, the chance of 16 or more successes out of 192, when the chance of one success in one attempt is one out of twenty, and the result of each attempt is independent of each other attempt, is approximately eight percent, at least according to this other slashdotter.
Quite the conicidence isn't it? The chances of both succeding in total and the total faliure rate being the same figure. Did this slashdot poster provide you with the calculation? Did he provide you with the time frame for success? Did he provide the statistics on why the "most likely number of times for them to have seen spikes that large was about ten but they saw sixteen"? Did he give you anything at all?
Yes, he did provide the calculation. No, it wasn't your simple inane "success" divided by "attempts". Yes, the near-similarity between the two is a coincidence. No, you saying witty things like "Quite the coincidence, isn't it?" doesn't change the fact that it's a coincidence.
Do you want me to calculate it for you, in painful detail? Well I'm not going to. But I'll tell you how to do it:
Here's a basic law of probability for you: If you have a test that has an P chance of success in any single attempt, and each attempt does not affect any other attempt, then the chance of getting exactly S successes in A attempts is equal to:
( A! * ( P ^ S ) * ( ( 1 - P ) ^ ( A - S ) ) / ( S! * ( A - S )! )
In this case, A is 192 (they tried 192 times), S is 16 (they succeeded 16 times), and P is 0.05 (the chance of a success in any particular attempt was one in twenty).
So plug those numbers into that formula, and you'll get the probability of seeing exactly 16 successes in 192 attempts. If you don't believe me, look it the fuck up before launching into another diatribe.
But that's not what we're interested in - we're interested in the probability of seeing 16 or more successes in 192 attempts.
So, you'll have to calculate that forumula a bunch of times, plugging in the numbers 16, 17, 18, and so forth, up to 192, for the number of successes. Then add up the results of each of those individual calculations.
I'm not about to do that. And I'm not going to guarantee that the number you get by it will be "eight percent". But I will guarantee two things:
(1) It's the correct formula;
(2) If you do get "eight percent", the fact that 16 / 192 is approximately the same is... wait for it... coincidental.
How far back do they go? Do they set the limit before they start looking? If not, and they're prepared to look back indefinitely, they'll always find a spike!
And yet, they don't always find a spike.
In which case, the rate of 'coincident' spikes will of course be significantly higher than random chance!
This is your "then" to your "if" of "if they always find a spike". Since that "if" turns out to be untrue, your "then" is irrelevant.
No, you are wrong. Instead of the chance of succeeding 16 times out of 192 being 8%, the OVERALL success rate is 8%(16/192 * 100 = 8.3%). That is an incredible Failure rate, almost one that could not be attribited to chance. To calculate the chance of sucess vs. failure, you would have to know the time frame in which they designate success, one figure which they do not seem willing to divulge.
You have no idea what you're talking about.
That 16 out of 192 was 16 times that they saw a spike which was large enough so that it should happen, on average, one time out of every twenty.
That is, out of those 192 times, the most likely number of times for them to have seen spikes that large was about ten.
But they saw sixteen.
The chance of seeing sixteen (or more) can be calculated, and it turns out (at least according to another slashdot poster) to be about eight percent.
That 16/192 is around eight percent is merely coincidence. Your calculation has nothing to do with anything at all.
No - the heart of their claim is that some guy off the street can use ESP to affect the RNG - regardless of distance.
No, that's the conclusion they've drawn from the heart of their claim.
The heart of their claim is that they've done certain things which probability says have a certain chance of being able to be done.
But here's the problem: Starting from a "major event" and working backwards, you will always find a major spike... so long as you look far enough.
Yes, that's true. However, they obviously don't simply just look and look and look further and further back from the event until they find a spike, as evidenced by the fact that most of the time, they don't find spikes.
Assuming that they look for some particular time range around the event - and I haven't been presented with a reason to believe they do otherwise - then the chance of there being a spike during that range is quantifiable. Their claim is that they have exceeded the quantifiable chance, significantly.
ere, they've got two fuzzy criteria -- What is a "major event"?
That's not really relevant. Assuming that their data is not just a fluke (and not incorrect or doctored), then at the very least, they've shown that they predict spikes at a rate significantly greater than random. They seem to be taking for granted that that is (or would be) because of the "major events", but that conclusion is not strictly within the purview of science, based upon the methodology they're using.
How far in advance?
Again, there's nothing that I've seen that suggests their answer to this question is necessarily "fuzzy". And if it's not fuzzy, then its effect is quantifiable.
And much like the canals of Mars, it seems likely that in the random data they are seeing something they want to see.
I think you're guilty, here, of something that a lot of posters in this thread are guilty of: You're assuming that they're not familiar with statistics.
They claim to have statistically quantified the chance of them successfully predicting spikes at the rate that they have, in fact, successfully predicted spikes. The claim being made is that that's a surprisingly low number.
(1) You're correct that you misunderstood me.
/.er are correct).
(2) And once again, you misunderstood me:
I said:
there was a coincident spike in significantly more cases than would have been predicted by random chance.
You replied:
Errr....Random chance doesn't predict anything.
Fine, fine. Let me be painfully explicit:
Define "success" as them saying "maybe there's a spike now", then checking the box, and finding that there is a spike of a certain statistical significance or higher.
Define "failure" as them saying "maybe there's a spike now", then checking the box, and finding that there is not a spike of that certain statistical significance or higher.
They have tried 192 times, "succeeded" 16 times, and "failed" 176 times.
The theory of probability can be used to predict that, if "success" or "failure" is random, your chance of "succeeding" that many times or more in that many attempts is eight percent (assuming calculations done by another
So they where looking for September 11th then? Or the tsunami? The article indicated that these events were the boxes major triumphs. If thats the case, why do they need the box at all?
It's not entirely clear to me what you're asking.
(1) They noticed that 9/11 happened.
(2) They thought to themselves, "This is a significantly major event that might cause a spike, if the proposed theory is real".
(3) As they always do when they think such things to themselves, they went back and checked the box.
(4) There was a coincident spike.
(5-8) Repeat (1-4) with "the tsunami" in place of "9/11".
(9-large) Repeat again, many more times, for other events, except in most cases, there was no coincident spike.
Yes, in most cases, there was no coincident spike. However - and this is the heart of their claim - there was a coincident spike in significantly more cases than would have been predicted by random chance.
The glaring error in their methodology is that they are Quantifying the un-quantifable. A significant world event is an opinion. What constitues a significant world event is in the eye of the beholder.
As pointed out several times, they don't look for spikes and then look for corresponding major events.
They look for events that they think might cause a spike (if the proposed effect is real), and then see if a spike occurred coincidentally.
So far, they've found such spikes, in this manner, at a rate significantly higher than random chance. Another poster calculated their chances of prediction at this level or higher at eight percent.
Whether them successfully doing something that has an eight percent chance of happening is something to get worked up about or not is up for debate. But your statement about their "glaring error" has nothing to do with anything.
It took me all of 5 seconds to find this article which pretty much debunks the entire project:
The article you link to debunks nothing.
Basically, the argument of the article is this:
(1) He showed me that, coincidentally with such-and-such an event, the RNGs strayed significantly from 50/50.
(2) I showed him that they also did the day before.
(3) I asked him what happened the day before.
(4) He said he didn't know.
(5) Ergo, what a load.
While that argument is seemingly cogent from a simplistic point of view, a little thought reveals that it's basically irrelevant. Specifically, the argument merely debunks a claim that is not being made in the first place.
These spikes happen all the time. No one denies that. Pointing out that - which is all the supposed "debunking" article does - is neither here nor there.
The claim being made is that the spikes happen more frequently than predicted by random chance when "major" events happen. This supposed "debunking" article does not address that claim at all. At all.
The project started with a scientist asking people to come to his lab and concentrate. That one person allegedly influenced the outcome of the random number generator. This would tend to suggest that your influence on the machine is directly related to your distance from it.
How so?
As far as I know, and as far as the article seems to mention, the test did not investigate any potential link between the distance and the outcome.
Yes, the person was in the same lab as the machine. But the test did not then go on to try with the guy in the same block, the same city, the same continent.
All it did was test "the guy thinking about influencing the machine" versus "the guy not thinking about influencing the machine".
To draw any conclusion, based upon that, about distance being involved is unscientific.
Basically, what they did is the following: They looked at the output of this device for "statistical anomalies", i.e. the .5% of the time it's output was out of the range it was expected to be in 99.5% of the time. Then, they looked for "important events" that occurred around that time. They don't have any way of choosing them other than that they happened when this occurred. Finally, they started looking for events *after* it happened, and found those as well. BOOM, it predicts the future.
That's actually utterly false.
As mentioned elsewhere in this thread - multiple times, in fact - they don't look for deviations then look for news that occurred near the time of the deviation.
They look for news that they think might cause deviations if this effect were real, and then look for deviations at the time of that news.
Seriously, they can't. Randi needs a totally controlled environment. The project uses a random generator and world events, both of which I don't think you can get into a controlled envrionment that would satisfy all.
That's not entirely true. The article mentioned and quoted several independent people - scientists, apparently - who claim that merely asking people to think about affecting the numbers generated by random number generators does, in fact, affect the numbers generated by random number generators.
That can easily be tested in a controlled environment. In fact, it's what these scientists in the article claim to have already done.
Read this book. You'll love it, trust me on this one.
Well, assuming you're a teenager.
What terms were violated, specifically?
I'm not (necessarily) doubting you - I'm mostly just curious.
Uh, that's accomplished, and then some, by instead getting used to a one-standard world.
I officially proclaim us at (or beyond) the point where we can say "screw people with Netscape 4.0 or IE 3 or whatever".
The existing differences between the rendering on the current versions of the main browsers (and most minor browsers too) are so trivial that a completely standards-compliant page can be made to look good in any of them, even if they might look slightly different in each.
IE misinterprets the box model? So what? With reasonably chosen values, things look fine in each. That fancy bevelled border that you want shows up as a plain old border in IE? Who cares? There are worse things than a plain border. And so forth.
A browser, email client, calender, and office suite do not an operating system make.
That minor point aside, a main problem with the whole XUL concept, in terms of extensibility, is the lack of good namespacing (or similar) capabilities in Javascript. If your extension uses a function called "getData", and mine uses a variable called "getData", guess what, there are problems. And nothing is going to tell you "hey, there's a conflict" - the underlying engine will happily call one of them. Maybe not the one you wanted. Global variables further complicate the issue.
Of course, today, this kind of problem is merely one that makes plugin developers scratch their heads and waste a lot of time, until finally they realize "Oh, goddamn, this programming environment is idiotic", and fix the immediate problem.
But if XUL has grand schemes on everything in the entire computer world, this issue has to be addressed.
Another issue is that the base applications don't always give names to widgets that you may want to modify. As far as I know, there is thereafter no way to safely extend those in a plugin - maybe its parent is named "blah", and you can say "modify blah's fourth widget child", but what if another plugin programmer inserted a widget before blah's (initial) fourth, thus making the one that you want to modify the fifth? Then you're modifying some random widget that you didn't intend to, with unpredictable and unintended consequences.
If you think C/C++ buffer overflows are bad, wait until XUL takes over the world. You'll have malware that hijacks function calls, malware that intentionally causes popular plugins to modify things that they weren't intended to modify, and so forth.
Wow, that was modded unexpectedly harshly.
How can the first post be modded as "redundant"?
Google cache is not particularly useful as a tool against slashdotting when the main objects of interest are the graphics and video stored on the slashdotted server.
No, the word I'm looking for is the word I used: "strangle".
Perhaps the poster that I was responding to was looking for "abdicate", but that's an entirely different question.
Resign?
Brits should really take matters into their own hands and kick him, and the rest of their supposed superiors, out.
In the words of Denis Diderot, "Mankind shall not be free until the last king is strangled in the entrails of the last priest."
It is "firmly pitched at the coder", but it has to explicitly state that it "assumes that you're not scared of mark-up"?
If you're "scared" of mere markup, I would hate to see your actual coding.
Movie review shows aren't allowed to tape their own clips from the movie. The only clips you see in a movie review show are those provided by the studios to review show at the sole discretion of the studios. The studios don't release the whole film to the production companies who make review shows, only the clips the movie companies want to be seen.
You're assuming we're talking about new movies. The post I responded to did not. He was also talking about ripping from DVDs.
The only reason (new) movie review shows only show the clips provided to them is because they're the only clips they have legal access to. If they had legal access to the rest of the movie - as they do when they are reviewing things that have newly come out on DVD - they could legally pick and choose from any portion.