It's more of a "testing is dropping it this many times" number; if the laptop fails within that many drops, it fails the test. It's the minimum for the machine to survive to be to spec. That said, apparently the actual test spec is for the packed condition, not the bare machine. So they weren't testing it appropriately, and it's built to be much wimpier than a toughbook, which is tested for a 90cm drop onto every corner, face, and side, onto plywood over steel over concrete, with the bare laptop, both open and closed, and both running and not.:)
The one I'm writing on (CF-27 toughbook) wouldn't flinch even if it were on. You'd not want to do it on a table you care about, though, as it'll leave gouges in the surface; there's no plastic padding on the corners. A friend of mine was giving a presentation on a CF-26, and when asked if that was a toughbook, unplugged the cable, closed the hatch over it, threw the laptop against the brick wall, running, picked it up, plugged it in, and finished the presentation. Of course, that did crack the screen, but it was decidedly an out-of-spec event.
I can't really say much for the newer models, but I'm typing this on a CF-27 toughbook that's dropped from my back (probably something like like 150cm; I carry it on a strap) onto a concrete floor; the paint scratched, and the floor chipped rather substantially. It takes a point impact to the center of the back of the screen, or being fully thrown into a brick wall to crack the screen, and the machine will still run afterwards, it'll just have a visible crack in the screen. Especially the older toughbooks are well beyond the military spec for such things.
Right on the nose; the heavier elements are produced in the actual supernova, when there is so much energy flying around that the reactions can take place, absorbing all the energy they want, and can produce the really heavy elements. Yes, they don't produce much - but then, there are tiny amounts of most elements relative to the amount of hydrogen and helium in the universe - just look at this planet - a lot of it is iron, and most of the rest is stuff lighter than iron. Fission plants are releasing the long-stored energy of supernovae:)
As a contrived example, take the sum of 1000 values of.2 in two digit scientific notation - you sum the first 50 or so with no problem, getting the correct answer of 1.0*10^1, and then you add the next.2, and the answer is 10.2, which is represented as... 1.0*10^1. Then you add the next, and... again, the value doesn't change. Now, if you are doing this in binary,.2 isn't even precisely representable with a finite-length string of bits - so you are _always_ getting this problem. The example above comes out off by a factor of 20! and the problem does appear in less extreme forms in non-contrived examples - with differences in how a number gets rounded, etc. because of the sequence of values being added and other such things that should not be affecting the result.
At which point it becomes a matter of the worker's own responsibility for being stupid with a dangerous machine. If the machine is supposed to only operate if it knows exactly where you are, and you willfully lie to it about where you are, then it's your own fault. My inclination would be to give him a warning, and then get him out of the factory if he persists - clearly he isn't concerned enough about his own life and limb to be there.
In all honesty, you need to be scared around it. Even if you have this system, you still need to be scared of the table saw. Just because the table saw is supposed to notice you and stop doesn't mean that it actually will; what if, say, some fleck of metal gets into the electronics, and shorts the mechanism holding that block back (haven't gotten to the article, but that's how I'd design such a thing - try to fail on the stop side), and when you trip into it moving too quickly it doesn't stop? While you shouldn't be scared silly by the machine, you should be scared to the point that you are always aware that being around this machine is inherently dangerous and always will be.
In heavy industry you are more likely to see a safety feature like "both of your hands must be in the boxes and you must press both buttons to activate the machine that'd make you into paste". I'd be surprised if there were still many machines requiring manual work while running; they're less efficient and harder to make safe. And to boot, the two-switch system is much, much simpler and more robust. Much more likely to fail in the nonfunctional direction than fail in the squish someone sort of way.
The reasons I'd see are cost, added complexity, and a reduced respect for the machine. These machines will very likely be more expensive, they are more complex than a standard saw and therefore more liable to failure (no, I won't cut that piece of material, it looks like skin...), and they won't neccessarily tear you to shreds, so you'll be less scared of them, and more likely to get sloppy about safety. Not overwhelming concerns, but concerns.
On the other hand, with no atmosphere, a mass driver (Wikipedia quotes a figure of 50% effeciency for some designs) could certainly reach the 2.4 km/s to leave the moon; it'd take about 11GJ to run a tonne of payload to anywhere on earth. Your average mediocre terrestrial nuke plant runs something like 1.5-2GW, so that'd give you about 7 seconds to generate the power for a 1-ton shot. That's a bit over 12 thousand tons a day capacity. There's still how to dump all that energy and all the energy from the drop on reentry, but worst case for that seems to be heavy heat shields and parachutes. Combine that with some way of getting rid of some energy in space (this is what ion drives are for, or the goofy electromagnetic-drag-line-in-the-earth's-field setup - small devices that give you a lot of change in energy very slowly), and you've also got a system for putting things in just about any earth orbit you want below the energy of the moon's orbit.
Not saying that it's actually worthwhile, but that part is thoroughly doable with relatively straightforward developments of tech we have now.
On the other hand, while it certainly doesn't make sense to drop port numbers, it should (assuming the huge supply changes pricing the way it ought to) make it a lot more practicable to have one machine do the work of multiple machines by having it actually present itself to the network as multiple machines, rather than pretending by looking at domain names in the protocol or by using nonstandard port numbers for all but one server.
Now, I've heard (and tried) that a sufficient dose of garlic will affect the scent of your skin/sweat/etc such that mosquitos no longer find you appealing. I actually tried this one year at boy scout camp - consumed a couple cloves of garlic raw a day, didn't bother with repellant. Didn't get bitten. Other people did.
Hardly a scientific study, that, but it certainly might be worth a study, and I'll be carrying on with it in the current insufficiently tested state...
The media player issue was phrased in the positive, not in the negative. Not "You must not bundle the Media Player with Windows", but rather "You Must provide a version of Windows that does not include the Media Player." More options to the consumer, not less. They can still sell bundled Windows and Media Player, they just need to sell only Windows, as well.
Well, i should RTFA next time; barring the technology used for that connectivity, this is exactly what the article proposes; use the cooperative model to distribute bandwidth, and then essentially purchase bandwidth "in bulk", at a POP.
So don't run it that way; run a co-op. That is, have your group of people who get a network set up in some fashion to distribute the purchased bandwidth (I'm actually thinking wireless for this, perhaps with modified wrt54gs-es to bridge to ethernet), and then they purchase portions of some bandwidth source that leads to the internet at large. If someone isn't buying network access, you could even let them access computers on the wireless network, but they don't have a place on the uplink. Of course, I'm not sure of the practicality of such a system.
Strictly speaking, if you got the sources in binary-only format with an offer of the source if you want it, then you are only obligated to refer them to the upstream provider. Section 3 c. Downloading an ISO image, burning it, and giving it to them is in binary format, and telling them to go download the sources from Mepis is explicitly good enough. On the other hand, if you were commercially distributing those discs, or had received them in non-binary format, then you'd need to make the sources available yourself.
I was not so much trying to wave this up and down as "Hey! this is a great idea! you should all do this!" as much as point out that not every broadband provider prohibits the sharing of network access without the purchase of some more expensive service to allow large numbers of users; there is the occasional one who rather promotes such sharing.
Of course this does not mean that you no longer have a responsibility to see that illegal activities do not happen on your network; so do they. You are being the ISP to the people you are sharing your link with, speakeasy is just helping. As for issues of available bandwidth, you could without too much difficulty invest in a routing device that can divide the available bandwidth in an intelligent manner, or provide you with preferential service.
It's mostly the not imposing artificial restrictions that are not based in the technological and legal situation through the use of a TOS that explicitly states not to reduce their direct customer base that is appealing.
An account with Fon either includes information about the wire going into your house to the router you are earning your pass with, or your billing information, both of which are pretty good at telling who you are.
Sorry, but while Fon routers don't encrypt their wireless links, they do require a login in order to give you a link to the internet. You don't need to log MAC addresses, and that won't work - (macchanger lets you pick any random mac address you want, or just generates a random mac addy for you). You just need to log in with your account at Fon to use one of these routers, which identifies you quite well, and probably involves communicating with their servers, which certainly can keep logs.
Hence the login, which includes paying if you aren't in the system, and which involves being a known person who is running their own AP if you are.
Fon does handle accountability. It's not just "Hey! I've got an open access point here! Have fun!" it's "this is a node in the network managed by this company. You have identified yourself with this company, so you are allowed to use this node in exchange for whichever of these return services is most convenient for you."
Er, not quite so much. That may be one of the more common positions that seem to be advocated by his books, but it is not by far the only one.
For instance, "starship troopers" is more about the nature of war and civic duty than about commerce. Also springing to mind are "Stranger in a strange land", which some have suggested was part of the origin of the whole "free love" movement. "The Moon is a harsh mistress" doesn't so much go for capitalistic propaganda as the assertion that paying for what you get is unavoidable, even in a society that tries to obscure it. "Beyond This Horizon" goes farther, with money being considered just a representation of a share in the production of the society.
All in all, my impression is that the primary purpose in the whole set is to get people thinking. Some of the books make Heinlein seem to come from one side; other books make it look like his actual opinion is completely on the other side.
The predominant impression I got from them was one of "relationships are important" and "human beings should think". I mean, one of the main themes in "starship troppers" was the idea of formulating a system of morals based on clear and precise reasoning from a set of axioms.
"No boobs or underage sex" - Somehow I don't think he was referring to Heinlein in any way shape or form... I mean, sure, the juveniles stayed a bit to the side of direct mention, but Moon, or Time Enough, or certainly Stranger?
I think that was contrasting them, not suggesting that they were the same. Although by a more useful definition of "underage" than "below-this-prescribed-number-of-years", I would not be surprised at that being not particularly present.
1kWh=3600000 J= 3.6MJ
Multiply that by about 100 million units, assuming most of them are plugged in all the time:
360 terrajoules
Your average nuke plant produces.6 gigajoules per second.
So, it'll take an average nuke plant 360*10^3/.6 seconds, or 600000 seconds.
600000/(60*60*24)=6.94 days.
Slightly under 2% of the yearly output of one of your average nuclear power plants.
Slashdot Mirror
It's more of a "testing is dropping it this many times" number; if the laptop fails within that many drops, it fails the test. It's the minimum for the machine to survive to be to spec. That said, apparently the actual test spec is for the packed condition, not the bare machine. So they weren't testing it appropriately, and it's built to be much wimpier than a toughbook, which is tested for a 90cm drop onto every corner, face, and side, onto plywood over steel over concrete, with the bare laptop, both open and closed, and both running and not. :)
The one I'm writing on (CF-27 toughbook) wouldn't flinch even if it were on. You'd not want to do it on a table you care about, though, as it'll leave gouges in the surface; there's no plastic padding on the corners. A friend of mine was giving a presentation on a CF-26, and when asked if that was a toughbook, unplugged the cable, closed the hatch over it, threw the laptop against the brick wall, running, picked it up, plugged it in, and finished the presentation. Of course, that did crack the screen, but it was decidedly an out-of-spec event.
I can't really say much for the newer models, but I'm typing this on a CF-27 toughbook that's dropped from my back (probably something like like 150cm; I carry it on a strap) onto a concrete floor; the paint scratched, and the floor chipped rather substantially. It takes a point impact to the center of the back of the screen, or being fully thrown into a brick wall to crack the screen, and the machine will still run afterwards, it'll just have a visible crack in the screen. Especially the older toughbooks are well beyond the military spec for such things.
Right on the nose; the heavier elements are produced in the actual supernova, when there is so much energy flying around that the reactions can take place, absorbing all the energy they want, and can produce the really heavy elements. Yes, they don't produce much - but then, there are tiny amounts of most elements relative to the amount of hydrogen and helium in the universe - just look at this planet - a lot of it is iron, and most of the rest is stuff lighter than iron. Fission plants are releasing the long-stored energy of supernovae :)
As a contrived example, take the sum of 1000 values of .2 in two digit scientific notation - you sum the first 50 or so with no problem, getting the correct answer of 1.0*10^1, and then you add the next .2, and the answer is 10.2, which is represented as... 1.0*10^1. Then you add the next, and... again, the value doesn't change. Now, if you are doing this in binary, .2 isn't even precisely representable with a finite-length string of bits - so you are _always_ getting this problem. The example above comes out off by a factor of 20! and the problem does appear in less extreme forms in non-contrived examples - with differences in how a number gets rounded, etc. because of the sequence of values being added and other such things that should not be affecting the result.
At which point it becomes a matter of the worker's own responsibility for being stupid with a dangerous machine. If the machine is supposed to only operate if it knows exactly where you are, and you willfully lie to it about where you are, then it's your own fault. My inclination would be to give him a warning, and then get him out of the factory if he persists - clearly he isn't concerned enough about his own life and limb to be there.
In all honesty, you need to be scared around it. Even if you have this system, you still need to be scared of the table saw. Just because the table saw is supposed to notice you and stop doesn't mean that it actually will; what if, say, some fleck of metal gets into the electronics, and shorts the mechanism holding that block back (haven't gotten to the article, but that's how I'd design such a thing - try to fail on the stop side), and when you trip into it moving too quickly it doesn't stop? While you shouldn't be scared silly by the machine, you should be scared to the point that you are always aware that being around this machine is inherently dangerous and always will be.
In heavy industry you are more likely to see a safety feature like "both of your hands must be in the boxes and you must press both buttons to activate the machine that'd make you into paste". I'd be surprised if there were still many machines requiring manual work while running; they're less efficient and harder to make safe. And to boot, the two-switch system is much, much simpler and more robust. Much more likely to fail in the nonfunctional direction than fail in the squish someone sort of way.
The reasons I'd see are cost, added complexity, and a reduced respect for the machine. These machines will very likely be more expensive, they are more complex than a standard saw and therefore more liable to failure (no, I won't cut that piece of material, it looks like skin...), and they won't neccessarily tear you to shreds, so you'll be less scared of them, and more likely to get sloppy about safety. Not overwhelming concerns, but concerns.
On the other hand, with no atmosphere, a mass driver (Wikipedia quotes a figure of 50% effeciency for some designs) could certainly reach the 2.4 km/s to leave the moon; it'd take about 11GJ to run a tonne of payload to anywhere on earth. Your average mediocre terrestrial nuke plant runs something like 1.5-2GW, so that'd give you about 7 seconds to generate the power for a 1-ton shot. That's a bit over 12 thousand tons a day capacity. There's still how to dump all that energy and all the energy from the drop on reentry, but worst case for that seems to be heavy heat shields and parachutes. Combine that with some way of getting rid of some energy in space (this is what ion drives are for, or the goofy electromagnetic-drag-line-in-the-earth's-field setup - small devices that give you a lot of change in energy very slowly), and you've also got a system for putting things in just about any earth orbit you want below the energy of the moon's orbit. Not saying that it's actually worthwhile, but that part is thoroughly doable with relatively straightforward developments of tech we have now.
On the other hand, while it certainly doesn't make sense to drop port numbers, it should (assuming the huge supply changes pricing the way it ought to) make it a lot more practicable to have one machine do the work of multiple machines by having it actually present itself to the network as multiple machines, rather than pretending by looking at domain names in the protocol or by using nonstandard port numbers for all but one server.
Hardly a scientific study, that, but it certainly might be worth a study, and I'll be carrying on with it in the current insufficiently tested state...
The media player issue was phrased in the positive, not in the negative. Not "You must not bundle the Media Player with Windows", but rather "You Must provide a version of Windows that does not include the Media Player." More options to the consumer, not less. They can still sell bundled Windows and Media Player, they just need to sell only Windows, as well.
Well, i should RTFA next time; barring the technology used for that connectivity, this is exactly what the article proposes; use the cooperative model to distribute bandwidth, and then essentially purchase bandwidth "in bulk", at a POP.
So don't run it that way; run a co-op. That is, have your group of people who get a network set up in some fashion to distribute the purchased bandwidth (I'm actually thinking wireless for this, perhaps with modified wrt54gs-es to bridge to ethernet), and then they purchase portions of some bandwidth source that leads to the internet at large. If someone isn't buying network access, you could even let them access computers on the wireless network, but they don't have a place on the uplink. Of course, I'm not sure of the practicality of such a system.
Strictly speaking, if you got the sources in binary-only format with an offer of the source if you want it, then you are only obligated to refer them to the upstream provider. Section 3 c. Downloading an ISO image, burning it, and giving it to them is in binary format, and telling them to go download the sources from Mepis is explicitly good enough. On the other hand, if you were commercially distributing those discs, or had received them in non-binary format, then you'd need to make the sources available yourself.
I was not so much trying to wave this up and down as "Hey! this is a great idea! you should all do this!" as much as point out that not every broadband provider prohibits the sharing of network access without the purchase of some more expensive service to allow large numbers of users; there is the occasional one who rather promotes such sharing. Of course this does not mean that you no longer have a responsibility to see that illegal activities do not happen on your network; so do they. You are being the ISP to the people you are sharing your link with, speakeasy is just helping. As for issues of available bandwidth, you could without too much difficulty invest in a routing device that can divide the available bandwidth in an intelligent manner, or provide you with preferential service. It's mostly the not imposing artificial restrictions that are not based in the technological and legal situation through the use of a TOS that explicitly states not to reduce their direct customer base that is appealing.
Which is to say, sharing the bandwidth you get through them = smaller bill.
An account with Fon either includes information about the wire going into your house to the router you are earning your pass with, or your billing information, both of which are pretty good at telling who you are.
Sorry, but while Fon routers don't encrypt their wireless links, they do require a login in order to give you a link to the internet. You don't need to log MAC addresses, and that won't work - (macchanger lets you pick any random mac address you want, or just generates a random mac addy for you). You just need to log in with your account at Fon to use one of these routers, which identifies you quite well, and probably involves communicating with their servers, which certainly can keep logs.
Fon does handle accountability. It's not just "Hey! I've got an open access point here! Have fun!" it's "this is a node in the network managed by this company. You have identified yourself with this company, so you are allowed to use this node in exchange for whichever of these return services is most convenient for you."
Er, not quite so much. That may be one of the more common positions that seem to be advocated by his books, but it is not by far the only one. For instance, "starship troopers" is more about the nature of war and civic duty than about commerce. Also springing to mind are "Stranger in a strange land", which some have suggested was part of the origin of the whole "free love" movement. "The Moon is a harsh mistress" doesn't so much go for capitalistic propaganda as the assertion that paying for what you get is unavoidable, even in a society that tries to obscure it. "Beyond This Horizon" goes farther, with money being considered just a representation of a share in the production of the society.
All in all, my impression is that the primary purpose in the whole set is to get people thinking. Some of the books make Heinlein seem to come from one side; other books make it look like his actual opinion is completely on the other side. The predominant impression I got from them was one of "relationships are important" and "human beings should think". I mean, one of the main themes in "starship troppers" was the idea of formulating a system of morals based on clear and precise reasoning from a set of axioms.
"No boobs or underage sex" - Somehow I don't think he was referring to Heinlein in any way shape or form... I mean, sure, the juveniles stayed a bit to the side of direct mention, but Moon, or Time Enough, or certainly Stranger? I think that was contrasting them, not suggesting that they were the same. Although by a more useful definition of "underage" than "below-this-prescribed-number-of-years", I would not be surprised at that being not particularly present.
1kWh=3600000 J= 3.6MJ Multiply that by about 100 million units, assuming most of them are plugged in all the time: 360 terrajoules Your average nuke plant produces .6 gigajoules per second.
So, it'll take an average nuke plant 360*10^3/.6 seconds, or 600000 seconds.
600000/(60*60*24)=6.94 days.
Slightly under 2% of the yearly output of one of your average nuclear power plants.
I'm sorry, my prefixes are broken. Re-running the numbers from scratch...