You should read the reports, particularly Feynman's. The technical people and the management were on diverging paths of bamboozlement.
The management, yes, overrode some technical people who said the risk was too high. They estimated the risk as tiny and represented it as tinier, as if the numbers were only meant to impress rather than indicating an actual mathematical quantity.
But they did it because some of the technical people said the risk was not too high. They deluded themselves to believe that it was OK to accept a higher risk because previous trials had not caused failure. This is the same syndrome that leads a roulette player to go bust, just because he won a few times in a row. True, not all the techs fell prey to it, but enough did that management got the impression that conflicting opinions of the risk existed (everyone on the technical side knew that the overall risk was around 1-2 percent, they differed about whether it was acceptable). Engineers raised the "success criterion" for tests to match their data when their tests failed, because the failure in that trial was not catastrophic, without understanding the root cause of the poor performance and the real odds of a catastrophic failure. As a result, components passed their tests in a state that initially would have cause a mission scrub due to an aggregate risk that was too high, and went into service without an assessment of true risk. The foam thing was another example of "it hasn't killed anyone yet, ignore it".
After a history of doing things this way, and a history of good luck, it became easier for management to ignore the warnings of higher risk and push forward. The message to take home is to treat every risk seriously, recognize that 1% chance of failure is not small - it means that you *will* fail one day if you keep trying.
Compression decreases the chance that a file will be corrupted by a random 1-byte or 1-block event (because it becomes a smaller target) but it increases the chance that you won't be able to guess the right way to fix it, and (with most algorithms) it ensures that the file beyond the point of the error (or sometimes the whole file) is useless.
If you're worried about loss, compress your data then use a separate parity system (like PAR2) to store redundant information about the compressed file, so you can reconstruct a certain number of bad bytes or blocks. Most of the parity programs have a feature that helps you determine the right degree of redundancy if you can define the degree of error you want to be able to tolerate. I like that better than hoping the compression algorithm makes the right assumptions.
Yes, this can increase the size of the data up to or beyond the original size, but you're in control of the size difference and by compressing and computing parity you can have better error tolerance than by not compressing at all (that is, you're more likely to be able to guess the missing information).
I too am concerned about these high-performance compression programs. I prefer open source so that I can do the math and be sure the algorithm is going to be 100% reversable. There's always some new miracle compression tool (anyone remember WEB?) that can compress anything but only uncompresses some things -- not the compression program I'd like to use:).
Hm, I think death penalties are pretty meaningless. Yah, it takes me twice as long to level, but I play to have fun where I am, not to get to the end.
If you get to the end, you're Done, in a MMORPG.
In city of heroes, the death penalty cuts experience in half for a certain amount of experience to earn. But not money. So one way to be a rich hero is to stay in debt.
Just a quick point -- origin specific armor is in the game -- it's a single-origin enhancement you get for most of those missions.
Before you holler "no! I mean something I can wear!" consider -- if it's in the game everone will have it. There's literally no way to distinguish your character from others in any MMORPG because anyone can copy your build. You can try, by making unpopular choices and with COH's wonderful costume variety, but fundamentally you'll be just as alike with a bit of armor or a weapon as you are now.
You can only distinguish yourself by personal expression -- your banter, your play style, whatever. The distinguishing marks have to come between chair and keyboard, anything in the game is just more of the same:).
BTW - I agree that most of the content is similar-ish in the abstract, but of all the MMORPG's I've played COH is the one with the most novelty over the shortest time, including gorgeous new maps, new quests, etc. (No, I don't EQ)
The story arcs and trials are a little more compelling than the regular missions, but they're becoming a way to accomplish the distinguishing marks you're looking for -- if they add many more you'll have to abandon some of them on your way, and the badges they offer will never be yours. It's still perfectly possible for others to have the same background, but it's one more set of combinations that will tend to make characters differ.
You're being under-paranoid. MS is offering a Royalty Free License for many of its technology APIs that clears you (the recipient) of MS copyright and patents for implementing standard protocols. MS requires that anyone distributing your code sign a copy of the license.
The license says that MS can tell others who the licensees are. Here's what MS is trying to do --
They shout "we're indemnifying our customers -- come and get a settlement!". UnderHandedCo finds out that they have some submarine patents, so they surface and serve MS. MS now can try to cross-license, or if UnderHandedCo is not in the software business, they just pay off. But with the envelope they deliver a list: "these people are implementing this standard that you have a patent on. You should sue them too."
Those licensees aren't indemnified -- they are implementing a protocol, not using MS software. So MS has just turned a pack of rabid attack dogs loose amongst its competition. This will raise the cost of building software (due to royalties), allowing MS to float by peacefully upon their cash reserve while FOSS and small businesses founder and sink.
I used to live in rural NC and we had packs running. If you didn't encounter any in rural Indiana or Pennsylvania, you may not have been rural enough:). The dogs seldom came to the road or houses.
I think that the PC gaming industry has showed that self-regulation can work. It's really in the sites' best interest to let interested adults find them easily, while enabling other venues to block them.
Poop. Pfiffle. Poppycock. Any plan that depends on self-rating may as well pack up and go home. You're ignoring human nature and the reach of legislation if you think otherwise.
One name: goatse.cx
There will always be someone who doesn't care, who wants to shock, who wants to reach the most eyeballs. And there will always be somewhere they can host.
Multiple rating services can cope with this, but of course that ignores the problem of looking up ratings on one or more remote databases for every URI you process including linked images and forwards. And nobody (not even google) has them all. Do you cut off the unknown URLs? Browse at 1 page a day? Pfiffle. Maybe for unsupervised browsing by a minor, but if that's the goal, why not send the kid to the library? They're less likely to run into trouble there, and the library probably has more resources than any service is likely to (accurately) rate.
I found that a lot of my RPM dependancy problems arose from packager decisions that were contrary to my preferences. With Gentoo I set my build preferences for which sound, graphic, processor, peripheral, etc. preferences I like and the ebuilds apply the correct compiler switches for me.
For instance, I have a few music and sound editing programs that can go KDE, Gnome or ALSA. When I was working with Redhat I had to extract sources, look up option settings, configure, make, repackage and install when the author's preference conflicted with mine (I could have just done make; make install, but I wanted to be able to remove it with rpm). So...with Gentoo I just emerge and all that happens automagically. I still know what I'm installing, and I can still stop and unpack the sources, but I don't need to as often.
And yes, I learned a bit more about system configuration with Gentoo, as the scripting is a little less involuted than dear old Redhat's. Not necessarily superior, but I find it easier to follow what happens when you try to start eth0.
My system works fine running X and compiling, oh, say openoffice or Gnome. It just chatters a lot. If your system isn't working well I suggest you look to your filesystems and memory.
Oops, reply to wrong thread. Ok, for this one, if your dream measurement is different from the definiton of blue, then you're not measuring blue. You're just seeing it. You're modelling the difference between reality and perception, rather than changing reality. And if you do measure it the same, you're following the internal logic that an external reality exists to cause your perception, even if you're only imagining the perception.
Finally, how do you know you're seeing 'blue'? how do you know it's not a shade of grey that you insist on remembering as 'blue'. Like when you dream you're on a double-date with your sister and her date, only they look like Goldie Hawn and Sean Connery, but you _know_ who it really is...
Re:Mathematics not universal?
on
The Golden Ratio
·
· Score: 3, Interesting
Um, no. I argue that an objective reality is separate from perception, and a couple of things about terms. Whether those things make sense depends on the degree to which my fever is affecting my perceptions...
There is a sloppy argument in the parent that perception = reality. That is demonstrably false, if one imagines the "impossible box" illusion, for instance. Besides, it doesn't matter: if there is not a reality, we're both insane figments of the reader's imagination, because only the reader is creating reality (hi reader! keep thinking about me for a while!). After all, I know it's not you and you know it's not me...
A more rigorous reading would be that the process of perception creates my personal experience of reality. Well duh. So reality is distinct from the perception model -- I know that the process of perception is imperfect (via repeated and sometimes painful demonstration), and does not in fact create an accurate model of the empirical reality that thwacks me in the nose when I misjudge a softball catch. That the information in the model is incomplete or contradictory is demonstrated whenever you discover an illusion. That there is a reality is demonstrated when it bonks you.
BUT... That we have unique understanding of 'blue' does not practically prevent us from conversing constructively about 'blue' and having high confidence that what I recognize as 'blue' will be recognized by you and any other capable person as 'blue'. To the degree that we are specific about the method of measuring 'blue' it becomes more likely that we can agree. It doesn't matter that your 'blue' is related to 'sour' in your mind, as long as we agree that it is the color of the sky (when the other would label the sky 'blue').
Aside - Edwin Land showed that color perception is largely relative - blue light is always around 470nm but... the perception of the hue of a color depends largely on the relative intensity of other wavelengths also present. He was able to produce full-color images from grey-scale filters in two different-color light sources.
So yes, 'blue' is an advanced concept that would have to be nailed down after months or years of discussion with the BEMs, possibly involving retinex algorithms to 'decide' if a thing is blue or not. There is, however, a 'blue' out there in reality to point at, and however they percieve it, we can explain to them that 'that' is 'blue'. Perception is relative, but reality is objective, for an agreed frame of reference.
A red stoplight indicates that you are not approaching it at a speed sufficiently close to the speed of light, but from the cop's point of view it's still red.
You're mistaking the model for the structure. It's the "Your Finger Stupid Mountain" problem all over gain. Or more properly, the difference between neurosis and psychosis -- one builds castles in the air, the other lives there.
Each person creates a model of reality based on their perceptions of reality. More accurately, the mind constructs a model from clues provided by the senses and then perceives the model. If your model and my model differ, then we disagree about reality, but we are probably not correct (either of us). Neither did either of us create the reality we disagree about. We created a model of it, each of us, and we're artists arguing over the accuracy our paintings while the subject gets dressed and moves along.
If individuals created reality, illusions would be impossible -- the individual would not be fooled, the impossibility would be true. Yet we enjoy magicians and artists that create illusions because they fool us into constructing inaccurate or contradictory models of reality, then reveal to us our errors (though perhaps not all of them).
So if you say that descriptions of reality are socially created, I'll say -- 'duh'. But if you say reality is socially created I'll ask you to be more precise in modelling your concept, or else to stop doing it so noisily.
If I see the color blue in a dream, where is the radiation?
In your dream silly. And if you carry a pair of dream-calipers and dream you're measuring the wavelength of the color, you'll find it reads around 470 nm...
By the way, if you are arguing that your dream is the same as reality, we have someone who should talk to you.
That'll last for about a week (ok a few years) but the DRM folks are pushing to make your tuner, your TV and even your surround-sound speakers respect DRM.
The point is that once we're all on digital receivers, prevention techniques like requiring authorized crypto signatures and broadcast flag recognition can keep you from playing your own stuff for others, unless you obtain someone else's permission.
In your example, public access may pass through a charge for adding DRM to your product. And the DRM may be mandatory as otherwise receiving sets may reject the signal.
Not that this would be popular, but the industry killed DAT.... they're not above shooting themselves in the foot again to maintain control.
SMP dual processor is not more reliable than single processor, it's 1/2 as reliable (either one goes down and the system crashes).
I'd be willing to bet that the 2 processor design was specced to create reliability, but it actually doubled the chance of failure due to implementation. It sounds like someone who knows something about computers was involved in the redesign. Not that I know anything....
Slashdot Mirror
You should read the reports, particularly Feynman's. The technical people and the management were on diverging paths of bamboozlement. The management, yes, overrode some technical people who said the risk was too high. They estimated the risk as tiny and represented it as tinier, as if the numbers were only meant to impress rather than indicating an actual mathematical quantity. But they did it because some of the technical people said the risk was not too high. They deluded themselves to believe that it was OK to accept a higher risk because previous trials had not caused failure. This is the same syndrome that leads a roulette player to go bust, just because he won a few times in a row. True, not all the techs fell prey to it, but enough did that management got the impression that conflicting opinions of the risk existed (everyone on the technical side knew that the overall risk was around 1-2 percent, they differed about whether it was acceptable). Engineers raised the "success criterion" for tests to match their data when their tests failed, because the failure in that trial was not catastrophic, without understanding the root cause of the poor performance and the real odds of a catastrophic failure. As a result, components passed their tests in a state that initially would have cause a mission scrub due to an aggregate risk that was too high, and went into service without an assessment of true risk. The foam thing was another example of "it hasn't killed anyone yet, ignore it". After a history of doing things this way, and a history of good luck, it became easier for management to ignore the warnings of higher risk and push forward. The message to take home is to treat every risk seriously, recognize that 1% chance of failure is not small - it means that you *will* fail one day if you keep trying.
Compression decreases the chance that a file will be corrupted by a random 1-byte or 1-block event (because it becomes a smaller target) but it increases the chance that you won't be able to guess the right way to fix it, and (with most algorithms) it ensures that the file beyond the point of the error (or sometimes the whole file) is useless.
:).
If you're worried about loss, compress your data then use a separate parity system (like PAR2) to store redundant information about the compressed file, so you can reconstruct a certain number of bad bytes or blocks. Most of the parity programs have a feature that helps you determine the right degree of redundancy if you can define the degree of error you want to be able to tolerate. I like that better than hoping the compression algorithm makes the right assumptions.
Yes, this can increase the size of the data up to or beyond the original size, but you're in control of the size difference and by compressing and computing parity you can have better error tolerance than by not compressing at all (that is, you're more likely to be able to guess the missing information).
I too am concerned about these high-performance compression programs. I prefer open source so that I can do the math and be sure the algorithm is going to be 100% reversable. There's always some new miracle compression tool (anyone remember WEB?) that can compress anything but only uncompresses some things -- not the compression program I'd like to use
Hm... high hopes. Why not just slice it up into good shapes for dropping?
Hm, I think death penalties are pretty meaningless. Yah, it takes me twice as long to level, but I play to have fun where I am, not to get to the end.
If you get to the end, you're Done, in a MMORPG.
In city of heroes, the death penalty cuts experience in half for a certain amount of experience to earn. But not money. So one way to be a rich hero is to stay in debt.
Just a quick point -- origin specific armor is in the game -- it's a single-origin enhancement you get for most of those missions. Before you holler "no! I mean something I can wear!" consider -- if it's in the game everone will have it. There's literally no way to distinguish your character from others in any MMORPG because anyone can copy your build. You can try, by making unpopular choices and with COH's wonderful costume variety, but fundamentally you'll be just as alike with a bit of armor or a weapon as you are now. You can only distinguish yourself by personal expression -- your banter, your play style, whatever. The distinguishing marks have to come between chair and keyboard, anything in the game is just more of the same :).
BTW - I agree that most of the content is similar-ish in the abstract, but of all the MMORPG's I've played COH is the one with the most novelty over the shortest time, including gorgeous new maps, new quests, etc. (No, I don't EQ)
The story arcs and trials are a little more compelling than the regular missions, but they're becoming a way to accomplish the distinguishing marks you're looking for -- if they add many more you'll have to abandon some of them on your way, and the badges they offer will never be yours. It's still perfectly possible for others to have the same background, but it's one more set of combinations that will tend to make characters differ.
You're being under-paranoid. MS is offering a Royalty Free License for many of its technology APIs that clears you (the recipient) of MS copyright and patents for implementing standard protocols. MS requires that anyone distributing your code sign a copy of the license.
The license says that MS can tell others who the licensees are. Here's what MS is trying to do --
They shout "we're indemnifying our customers -- come and get a settlement!". UnderHandedCo finds out that they have some submarine patents, so they surface and serve MS. MS now can try to cross-license, or if UnderHandedCo is not in the software business, they just pay off. But with the envelope they deliver a list: "these people are implementing this standard that you have a patent on. You should sue them too."
Those licensees aren't indemnified -- they are implementing a protocol, not using MS software. So MS has just turned a pack of rabid attack dogs loose amongst its competition. This will raise the cost of building software (due to royalties), allowing MS to float by peacefully upon their cash reserve while FOSS and small businesses founder and sink.
They want to do it because:
a) they have a patent warchest to make cross-license deals which don't cost them a dime and let their products off the hook.
b) they have a $30bn warchest for those that can't be threatened into a cross-license
c) Linux, OpenOffice, Apache and other free projects don't have these advantages.
They're trying to increase the cost of doing business for all programmers by saying "here's a money source, do you have a patent we can license?".
We're still waiting for that. But it's closer now, after about 16 months.
So if you're just getting around to it, buy the Garmin Gpsmap 60cs -- with compass and altimeter :)...
I used to live in rural NC and we had packs running. If you didn't encounter any in rural Indiana or Pennsylvania, you may not have been rural enough :). The dogs seldom came to the road or houses.
Poop. Pfiffle. Poppycock. Any plan that depends on self-rating may as well pack up and go home. You're ignoring human nature and the reach of legislation if you think otherwise.
One name: goatse.cx
There will always be someone who doesn't care, who wants to shock, who wants to reach the most eyeballs. And there will always be somewhere they can host.
Multiple rating services can cope with this, but of course that ignores the problem of looking up ratings on one or more remote databases for every URI you process including linked images and forwards. And nobody (not even google) has them all. Do you cut off the unknown URLs? Browse at 1 page a day? Pfiffle. Maybe for unsupervised browsing by a minor, but if that's the goal, why not send the kid to the library? They're less likely to run into trouble there, and the library probably has more resources than any service is likely to (accurately) rate.
Don't suck too many of those starch foam noodles down, they often contain anti-fungals to prevent stinky packages...
For instance, I have a few music and sound editing programs that can go KDE, Gnome or ALSA. When I was working with Redhat I had to extract sources, look up option settings, configure, make, repackage and install when the author's preference conflicted with mine (I could have just done make; make install, but I wanted to be able to remove it with rpm). So...with Gentoo I just emerge and all that happens automagically. I still know what I'm installing, and I can still stop and unpack the sources, but I don't need to as often.
And yes, I learned a bit more about system configuration with Gentoo, as the scripting is a little less involuted than dear old Redhat's. Not necessarily superior, but I find it easier to follow what happens when you try to start eth0.
My system works fine running X and compiling, oh, say openoffice or Gnome. It just chatters a lot. If your system isn't working well I suggest you look to your filesystems and memory.
Oh,no, at least one hollywood type found it. They made it into Soldier
Finally, how do you know you're seeing 'blue'? how do you know it's not a shade of grey that you insist on remembering as 'blue'. Like when you dream you're on a double-date with your sister and her date, only they look like Goldie Hawn and Sean Connery, but you _know_ who it really is...
There is a sloppy argument in the parent that perception = reality. That is demonstrably false, if one imagines the "impossible box" illusion, for instance. Besides, it doesn't matter: if there is not a reality, we're both insane figments of the reader's imagination, because only the reader is creating reality (hi reader! keep thinking about me for a while!). After all, I know it's not you and you know it's not me...
A more rigorous reading would be that the process of perception creates my personal experience of reality. Well duh. So reality is distinct from the perception model -- I know that the process of perception is imperfect (via repeated and sometimes painful demonstration), and does not in fact create an accurate model of the empirical reality that thwacks me in the nose when I misjudge a softball catch. That the information in the model is incomplete or contradictory is demonstrated whenever you discover an illusion. That there is a reality is demonstrated when it bonks you.
BUT... That we have unique understanding of 'blue' does not practically prevent us from conversing constructively about 'blue' and having high confidence that what I recognize as 'blue' will be recognized by you and any other capable person as 'blue'. To the degree that we are specific about the method of measuring 'blue' it becomes more likely that we can agree. It doesn't matter that your 'blue' is related to 'sour' in your mind, as long as we agree that it is the color of the sky (when the other would label the sky 'blue').
Aside - Edwin Land showed that color perception is largely relative - blue light is always around 470nm but... the perception of the hue of a color depends largely on the relative intensity of other wavelengths also present. He was able to produce full-color images from grey-scale filters in two different-color light sources.
So yes, 'blue' is an advanced concept that would have to be nailed down after months or years of discussion with the BEMs, possibly involving retinex algorithms to 'decide' if a thing is blue or not. There is, however, a 'blue' out there in reality to point at, and however they percieve it, we can explain to them that 'that' is 'blue'. Perception is relative, but reality is objective, for an agreed frame of reference.
A red stoplight indicates that you are not approaching it at a speed sufficiently close to the speed of light, but from the cop's point of view it's still red.
Each person creates a model of reality based on their perceptions of reality. More accurately, the mind constructs a model from clues provided by the senses and then perceives the model. If your model and my model differ, then we disagree about reality, but we are probably not correct (either of us). Neither did either of us create the reality we disagree about. We created a model of it, each of us, and we're artists arguing over the accuracy our paintings while the subject gets dressed and moves along.
If individuals created reality, illusions would be impossible -- the individual would not be fooled, the impossibility would be true. Yet we enjoy magicians and artists that create illusions because they fool us into constructing inaccurate or contradictory models of reality, then reveal to us our errors (though perhaps not all of them).
So if you say that descriptions of reality are socially created, I'll say -- 'duh'. But if you say reality is socially created I'll ask you to be more precise in modelling your concept, or else to stop doing it so noisily.
No, an egomaniacal idiot.
In your dream silly. And if you carry a pair of dream-calipers and dream you're measuring the wavelength of the color, you'll find it reads around 470 nm... By the way, if you are arguing that your dream is the same as reality, we have someone who should talk to you.
Odd, because I can't stand it unless I'm watching Letterman late at night -- he's just not as much fun in daylight (or when I'm not sleepy).
That'll last for about a week (ok a few years) but the DRM folks are pushing to make your tuner, your TV and even your surround-sound speakers respect DRM.
The point is that once we're all on digital receivers, prevention techniques like requiring authorized crypto signatures and broadcast flag recognition can keep you from playing your own stuff for others, unless you obtain someone else's permission. In your example, public access may pass through a charge for adding DRM to your product. And the DRM may be mandatory as otherwise receiving sets may reject the signal. Not that this would be popular, but the industry killed DAT.... they're not above shooting themselves in the foot again to maintain control.
I'm not sure I want to strap my limbs to hydraulically actuated levers that may or may not be capable of exceeding my range of motion....
SMP dual processor is not more reliable than single processor, it's 1/2 as reliable (either one goes down and the system crashes). I'd be willing to bet that the 2 processor design was specced to create reliability, but it actually doubled the chance of failure due to implementation. It sounds like someone who knows something about computers was involved in the redesign. Not that I know anything....