Chernobyl could have been built much more safely than Chernobyl (was built). But it cost less to build it as they did.
This particular oil rig could very likely have been built/operated more safely than it was. But who'll make BP do that?
Similarly, oil pipelines can be very safe, but they have been operated very unsafely, with maintenance neglected until accidents happen. It turns out that it's cheaper that way, lawsuits and all.
It's not a matter of what "we" can do. It's a matter of what government will actively regulate business to do. Business doesn't like regulation, and they often have more influence on lawmaking than "we" do. As long as no one pays much attention, they get their way.
> But in the next breath, the authors suggest folding away the values-list and taking a job (any job) that will pay the bills.
You'd think that they'd relate this to the bootstrapping process: A fancy filesystem (desired job) that's well-thought out is nice and good, but when you've got no OS ($) loaded, you need to get that code loaded by any means possible first, such as reading in the first few blocks off the boot drive (taking any job you can get). Once you've got a good base of code loaded, you can approach the task of loading data the right way.
Say what you want about Jobs, but one thing is clear: he's a businessman first and foremost. He knows how to make money. Regardless of what he may say about Flash, the decision to keep it off certain Apple products is a business decision that is aimed at making Apple more profitable.
Indeed, latency is big issue that makes remote operation difficult. We notice this all the time with the video-conferencing system we use that lets us aim the remote camera: push the left arrow, and the remote camera moves left a bit, and eventually you see the result on video (after a 1-2 second delay). If you hold down the left arrow and only let go when you see what you want to see on the feedback video, then you'll way overshoot, since once you let go, the camera keeps moving for another second or so. If you do this enough, you "learn" the latency, and learn exactly how early to stop the input to get the right feedback. Still, it's much trickier than when feedback is nearly instant.
The hospital should be doing many (IT-related) things. It seems like those in charge of the hospital's network security don't really have a clue, and they imagine that just installing encryption on all connected machines will solve a bunch of security problems. It doesn't really, and if done improperly, solves nothing at all and only creates new problems.
The first thing the hospital needs to do is to set up proper zones of security, such that machines with really confidential data are mostly sealed off from the rest of the network and the internet at large. There should be strict firewalls between those zones. Anyone who brings in their own computer should not have access to confidential data without setting up properly secured access.
I imagine though that there's plenty of opportunities for abuse of data security; for instance, employees might want to evaluate patient data just by emailing themselves a copy (or having copied it by other means, such as USB stick). This is where things get sticky, for once the cat is out of the bag, there's no telling where he'll go. I imagine that this is the problem the hospital is trying to solve, just by pretending that encryption will create a secure bag around any machine/transaction that they might be held responsible for.
Hmm, the more I think about it, the more I think it's likely that this is indeed just a big case of CYA. The hospital wants to avoid the legal charge that it didn't do everything possible to secure patients' private data. I suppose they could try to take it even one step further: require that anyone that employees send email to secure their computers as well.
I think folks are misunderstanding his misunderstanding.
It's obvious that if you connect to a single tracker, you get a list of peers who are downloading what that tracker has to offer. But that's not a picture of all the Internet's P2P traffic. It's just one data point.
What's not obvious is how you can start from zero information and then figure out who's downloading what across the whole internet. Do you do a google search for all torrent files, then try to figure out all the active trackers, then figure out all the peers? That seems like a huge task, and it's not exactly a real-time picture of who's downloading what so much as a collection of samples.
I peeked at the papers, and they start by going to the biggest providers of torrent files (The Pirate Bay, Mininova, etc.). Right away, this tells you that they don't see every single bit of P2P traffic. What they do reveal that isn't obvious to most people is that you can query a tracker not just about a particular torrent download, but about ALL the torrents that a given tracker tracks. This cuts down the work significantly.
In addition, those sites offer a list of brand new torrents, and if you hook up to one of those right away, you'll get a picture of who (which IPs) is injecting new content into the network. This is how they figure out who's likely to be doing that. Figuring out the big downloaders is more complex.
In any case, monitoring the P2P traffic is still a huge task, doesn't give a *complete* picture of all P2P traffic, and isn't as simple as making a single request.
Consider yourself lucky. I prefer neatly folded and stacked, whereas my wife prefers the cram-it-in approach.
As far as thirds of halves, I find I have to use both to fit larger things into their designated* spaces. My usual approach is to go halves until I realize that it won't fit as is and I can't fold it anymore due to thickness, and then I must back off a fold and go thirds.
* Ah, yes, I also prefer to put things in designated spaces (like with like), whereas my wife is less particular, except when *my* stuff is in the wrong place (which, admittedly, is not a rare occurrence). We get along by not complaining too much about the small stuff.
Time is an artifact of TV programming. For instance, dinner time always occurs around the evening news, while bed time is soon after the late show. Breakfast is when that silly morning news stuff happens, unless it's Saturday, when it coincides with cartoons.
Instead of shifting school an hour later, it would be interesting to see what happens if TV schedules were shifted an hour earlier.
It's not really possible to asses that. The article really doesn't have much to say about Linux, so much as it was about all the crufty patchwork of multiple systems they were using before. There's a big cost associated with continuing to use the current kludges, though it is difficult to assign hard numbers to, since they come in the form of lost opportunities and inefficiency spread throughout the whole organization.
Moving to any modern, unified system, whether based around Microsoft software or OSS, is a tremendous task for a big organization like that. And without a parallel universe (that made the other choice) to compare to, you cannot really say which choice was better. You can only guess. Sure, you can try to make an educated guess by trying to figure out how much of the legacy applications will still work on the new system without changes, but until you try to actually do that work, you won't know how wrong you were. [99% compatible is worthless if you were depending upon the 1% of things that don't work.]
That must have been a Dolby 3D projection. The Dolby system is based around color notch filters that only pass narrow bands of wavelengths around slightly different primary colors for each eye. Because the primary colors for each eye are slightly different, you might notice an effect such as you describe.
There's a more advanced system that passes multiple color bands for each primary that end up looking the same when viewed together, but I guess Dolby isn't using that yet. Perhaps it makes the glasses even more expensive.
Well, I think the real question in this case should be what is the topology of the shape in question (the human body)? Isn't the so-called "cavity" really just a long tube connecting two openings to the outer surface? If that be the only set of connected openings, then the body would be homeomorphic to a torus.
However, there's a complex set of connected openings in the head: 2 nostrils, 2 tear ducts, and the mouth all connect to each other inside. I don't know what this is referred to as, topologically. Perhaps someone can help me out here. I'm guessing it's a quad-torus, and combined with the hole above makes the total a quintuple-torus?
We do, of course, assume that no other piercings have been made.
In situations where I'm on the highway and I need to slow down quickly, I tap my brake pedal several times while slowing down, and tap it again a few times if I see lights in my rear-view mirror. Drivers might not pay enough attention to steady brake lights, but they'll notice flashing brake lights and hopefully pay more attention to what's happening in front of them. I myself have been caught by surprise, going down the highway at speed, coming around a bend, seeing the brake lights of the car in front and thinking he's just slowing for the curve, then realizing suddenly that he's slowing to a stop behind a line of stopped cars.
The Dolby technique is based on color notch filters. The lenses for each eye pass red, green, and blue light, but only at very specific wavelengths.
The basic form of this technique can lead to the colors looking a bit off, since the primary colors seen by each eye are slightly different.
However, a more advanced technique passes through multiple wavelengths for each primary so that the blended primaries appear to match each other. For instance, imagine that one eye passes light at 650 nm for the red primary, while the other passes light around 640 nm and 660 nm for red. Averaged out, both eyes appear to see red at 650 nm. Something similar is done for green and blue.
I don't know if Dolby has licensed the more advanced technique or not.
I'm aware of that. What I meant to say is that if you're watching the 2D version of a movie that was filmed in 3D, then you're seeing the same thing as if you went to the 3D movie and just looked with one eye through the glasses. Of course, it may be cheaper just to go see the 2D projection.
My comments were mainly to correct his notion of "misalignment" and color filters, as well as to address the notion that 3D glasses make the movie "too dark".
First off, the alignment is correct in 3D: there's a different image for each eye. You should never watch 3D movies without the glasses, even if you only have one good eye. Well, I suppose it's okay if you're blind. If you're watching a 3D movie in 2D, you're probably just watching one eye's view.
Second, the glasses' lenses aren't different colors, at least for any decent 3D method. They filter out the light so that the correct image goes to each eye. Some do it using different polarizations of light. Some do it using color notch filters (they pass all 3 primary colors, but only at specific wavelengths). And some do it by using LCD shutters to block out the image completely as each eye's image is shown alternately.
If you watch a 3D movie without the glasses, not only do you see 2 slightly different images superimposed on each other, but it's also twice as bright as it should be.
Slashdot Mirror
Reee-yull bad!
Chernobyl could have been built much more safely than Chernobyl (was built). But it cost less to build it as they did.
This particular oil rig could very likely have been built/operated more safely than it was. But who'll make BP do that?
Similarly, oil pipelines can be very safe, but they have been operated very unsafely, with maintenance neglected until accidents happen. It turns out that it's cheaper that way, lawsuits and all.
It's not a matter of what "we" can do. It's a matter of what government will actively regulate business to do. Business doesn't like regulation, and they often have more influence on lawmaking than "we" do. As long as no one pays much attention, they get their way.
And there's this nice meme from a couple years ago:
http://www.google.com/search?q=ken+lee
http://www.youtube.com/watch?v=_RgL2MKfWTo
"Time flies like an arrow; fruit flies like a banana."
> But in the next breath, the authors suggest folding away the values-list and taking a job (any job) that will pay the bills.
You'd think that they'd relate this to the bootstrapping process: A fancy filesystem (desired job) that's well-thought out is nice and good, but when you've got no OS ($) loaded, you need to get that code loaded by any means possible first, such as reading in the first few blocks off the boot drive (taking any job you can get). Once you've got a good base of code loaded, you can approach the task of loading data the right way.
Say what you want about Jobs, but one thing is clear: he's a businessman first and foremost. He knows how to make money. Regardless of what he may say about Flash, the decision to keep it off certain Apple products is a business decision that is aimed at making Apple more profitable.
Indeed, latency is big issue that makes remote operation difficult. We notice this all the time with the video-conferencing system we use that lets us aim the remote camera: push the left arrow, and the remote camera moves left a bit, and eventually you see the result on video (after a 1-2 second delay). If you hold down the left arrow and only let go when you see what you want to see on the feedback video, then you'll way overshoot, since once you let go, the camera keeps moving for another second or so. If you do this enough, you "learn" the latency, and learn exactly how early to stop the input to get the right feedback. Still, it's much trickier than when feedback is nearly instant.
The hospital should be doing many (IT-related) things. It seems like those in charge of the hospital's network security don't really have a clue, and they imagine that just installing encryption on all connected machines will solve a bunch of security problems. It doesn't really, and if done improperly, solves nothing at all and only creates new problems.
The first thing the hospital needs to do is to set up proper zones of security, such that machines with really confidential data are mostly sealed off from the rest of the network and the internet at large. There should be strict firewalls between those zones. Anyone who brings in their own computer should not have access to confidential data without setting up properly secured access.
I imagine though that there's plenty of opportunities for abuse of data security; for instance, employees might want to evaluate patient data just by emailing themselves a copy (or having copied it by other means, such as USB stick). This is where things get sticky, for once the cat is out of the bag, there's no telling where he'll go. I imagine that this is the problem the hospital is trying to solve, just by pretending that encryption will create a secure bag around any machine/transaction that they might be held responsible for.
Hmm, the more I think about it, the more I think it's likely that this is indeed just a big case of CYA. The hospital wants to avoid the legal charge that it didn't do everything possible to secure patients' private data. I suppose they could try to take it even one step further: require that anyone that employees send email to secure their computers as well.
I think folks are misunderstanding his misunderstanding.
It's obvious that if you connect to a single tracker, you get a list of peers who are downloading what that tracker has to offer. But that's not a picture of all the Internet's P2P traffic. It's just one data point.
What's not obvious is how you can start from zero information and then figure out who's downloading what across the whole internet. Do you do a google search for all torrent files, then try to figure out all the active trackers, then figure out all the peers? That seems like a huge task, and it's not exactly a real-time picture of who's downloading what so much as a collection of samples.
I peeked at the papers, and they start by going to the biggest providers of torrent files (The Pirate Bay, Mininova, etc.). Right away, this tells you that they don't see every single bit of P2P traffic. What they do reveal that isn't obvious to most people is that you can query a tracker not just about a particular torrent download, but about ALL the torrents that a given tracker tracks. This cuts down the work significantly.
In addition, those sites offer a list of brand new torrents, and if you hook up to one of those right away, you'll get a picture of who (which IPs) is injecting new content into the network. This is how they figure out who's likely to be doing that. Figuring out the big downloaders is more complex.
In any case, monitoring the P2P traffic is still a huge task, doesn't give a *complete* picture of all P2P traffic, and isn't as simple as making a single request.
Okay, now get busy hacking it so that it can be even more useful and open than it is now. Then I'll buy one.
Consider yourself lucky. I prefer neatly folded and stacked, whereas my wife prefers the cram-it-in approach.
As far as thirds of halves, I find I have to use both to fit larger things into their designated* spaces. My usual approach is to go halves until I realize that it won't fit as is and I can't fold it anymore due to thickness, and then I must back off a fold and go thirds.
* Ah, yes, I also prefer to put things in designated spaces (like with like), whereas my wife is less particular, except when *my* stuff is in the wrong place (which, admittedly, is not a rare occurrence). We get along by not complaining too much about the small stuff.
But that's the beauty of technology: in a couple of generations, it'll be 50 times faster!
Ok, maybe Moore's "Law" doesn't apply to everything technical...
You'd have a splendid wife if she could program robots like that!
Time is an artifact of TV programming. For instance, dinner time always occurs around the evening news, while bed time is soon after the late show. Breakfast is when that silly morning news stuff happens, unless it's Saturday, when it coincides with cartoons.
Instead of shifting school an hour later, it would be interesting to see what happens if TV schedules were shifted an hour earlier.
da-dum, da-da da-da da-dum!
It seems likely as it's really hard to breed a bunch of sterile bugs! :-)
It's not really possible to asses that. The article really doesn't have much to say about Linux, so much as it was about all the crufty patchwork of multiple systems they were using before. There's a big cost associated with continuing to use the current kludges, though it is difficult to assign hard numbers to, since they come in the form of lost opportunities and inefficiency spread throughout the whole organization.
Moving to any modern, unified system, whether based around Microsoft software or OSS, is a tremendous task for a big organization like that. And without a parallel universe (that made the other choice) to compare to, you cannot really say which choice was better. You can only guess. Sure, you can try to make an educated guess by trying to figure out how much of the legacy applications will still work on the new system without changes, but until you try to actually do that work, you won't know how wrong you were. [99% compatible is worthless if you were depending upon the 1% of things that don't work.]
That must have been a Dolby 3D projection. The Dolby system is based around color notch filters that only pass narrow bands of wavelengths around slightly different primary colors for each eye. Because the primary colors for each eye are slightly different, you might notice an effect such as you describe.
There's a more advanced system that passes multiple color bands for each primary that end up looking the same when viewed together, but I guess Dolby isn't using that yet. Perhaps it makes the glasses even more expensive.
Well, I think the real question in this case should be what is the topology of the shape in question (the human body)? Isn't the so-called "cavity" really just a long tube connecting two openings to the outer surface? If that be the only set of connected openings, then the body would be homeomorphic to a torus.
However, there's a complex set of connected openings in the head: 2 nostrils, 2 tear ducts, and the mouth all connect to each other inside. I don't know what this is referred to as, topologically. Perhaps someone can help me out here. I'm guessing it's a quad-torus, and combined with the hole above makes the total a quintuple-torus?
We do, of course, assume that no other piercings have been made.
These days, it seems you need a million dollars just to get by...
In situations where I'm on the highway and I need to slow down quickly, I tap my brake pedal several times while slowing down, and tap it again a few times if I see lights in my rear-view mirror. Drivers might not pay enough attention to steady brake lights, but they'll notice flashing brake lights and hopefully pay more attention to what's happening in front of them. I myself have been caught by surprise, going down the highway at speed, coming around a bend, seeing the brake lights of the car in front and thinking he's just slowing for the curve, then realizing suddenly that he's slowing to a stop behind a line of stopped cars.
The Dolby technique is based on color notch filters. The lenses for each eye pass red, green, and blue light, but only at very specific wavelengths.
The basic form of this technique can lead to the colors looking a bit off, since the primary colors seen by each eye are slightly different.
However, a more advanced technique passes through multiple wavelengths for each primary so that the blended primaries appear to match each other. For instance, imagine that one eye passes light at 650 nm for the red primary, while the other passes light around 640 nm and 660 nm for red. Averaged out, both eyes appear to see red at 650 nm. Something similar is done for green and blue.
I don't know if Dolby has licensed the more advanced technique or not.
I'm aware of that. What I meant to say is that if you're watching the 2D version of a movie that was filmed in 3D, then you're seeing the same thing as if you went to the 3D movie and just looked with one eye through the glasses. Of course, it may be cheaper just to go see the 2D projection.
My comments were mainly to correct his notion of "misalignment" and color filters, as well as to address the notion that 3D glasses make the movie "too dark".
First off, the alignment is correct in 3D: there's a different image for each eye. You should never watch 3D movies without the glasses, even if you only have one good eye. Well, I suppose it's okay if you're blind. If you're watching a 3D movie in 2D, you're probably just watching one eye's view.
Second, the glasses' lenses aren't different colors, at least for any decent 3D method. They filter out the light so that the correct image goes to each eye. Some do it using different polarizations of light. Some do it using color notch filters (they pass all 3 primary colors, but only at specific wavelengths). And some do it by using LCD shutters to block out the image completely as each eye's image is shown alternately.
If you watch a 3D movie without the glasses, not only do you see 2 slightly different images superimposed on each other, but it's also twice as bright as it should be.
A discussion is more than trying to show how a previous commenter is wrong.
Sometimes it's okay to just add information that has some relevance that others might find interesting.
mm was just providing a historical context for using GP's to do computation, which is what the original post is about.