Yes, the unit is the same (obviously, as they are both expressions for energy), but I interpreted the grandparent (or whatever it is) as drawing more far-reaching conclusions from the similarity of the two expressions. Rereading it now I'm not sure what he was trying to say.
SUV's are usually much more rigid, so while most modern cars are designed to crumple in a way as to minimize the decelerating force on the passengers, the entire SUV and its occupants will decelerate at essentially the same rate. While someone travelling in an ordinary car is 3-4 times more likely to die from crashing into a SUV than from crashing into another car, the SUV driver is also at a greater risk than a car driver. This is mostly because SUV's often have poor safety systems, eg no whiplash protection, bad handling, no anti-skid systems and so on.
A majority of all traffic deaths are from single vehicle accidents, and here the SUV's are much more dangerous than ordinary cars. Again because of rigidity, but also as an effect of their greater mass.
So, to sum it up, the SUV drivers are the much safer party in SUV-car collisions, as they increase the risk of other drivers more than the risk to themselves, but all in all run a greater risk of dying in a car wreck.
(The numbers in this post come from the Swedish insurance agency Folksam, http://www.folksam.se/. Not a promotion, simply a source reference)
"This incidentally does make it in the same class as E=m*c^2 (due to magnitude/exponentiation) [...]"
No, it doesn't work that way, as c is a constant. E=mc^2 scales linearly with mass and does not include speed, as it deals with the energy of mass at rest. E=½mv^2 scales linearly with mass and squarely with speed.
(What is the rationale behind slashcode removing ² - or ² - from posts?)
I have noticed that a couple of my users thouroughly enjoy the sight of GL screensavers (using xscreensaver), and they have never reported any lockups. I remind them regularly to report any and all annoyances with their computers, and they're usually pretty good at dong so, so I am fairly confident that there has been no lockups. I am currently running Xorg 6.8.2.
I admin a score of machines using Sapphire 9200 cards, all running with the Xorg driver. The machines are used daily (I am posting this from one of them) and I have yet to see a single problem with the driver. Granted, OpenGL is mostly used for screen savers on these boxes, but still. In my experience the drivers are rock solid.
Well, it wasn't entirely meaningless, as you understood exactly what I meant:)
I didn't intend it as a great revelation to solve all our security woes, but merely as a reminder of what the original problem we are trying to solve is. One should always keep the original objective in mind when attacking sub-problems. It is all too easy to get lost in the details and go against it.
"make a perl script in/usr/local/sbin; chown it root:bin; chmod it go-x. Users can still execute the sucker."
One has to remember that a program is just a sequence of instructions written down in a convenient form. Someone with access to a general purpose system can feed it general instructions. If your particular set of instructions are readable to him (which has been assumed so far in this discussion), he can feed them to the computer. He can do so with instructions he read on the Net, or on the back of a milk carton.
Perhaps what you are suggesting is that perl be configured so that it will only run when invoked from the script interpretor (i.e. when it is mentioned in a #!-statement in an executable text file). This would perhaps be an extra hoop to jump through, but would also greatly limit the usefullness of perl. I use perl -e 'something' all the time.
I don't consider this a design flaw, quite the opposite. Any general purpose operating system can always be told to execute arbitrary commands, that is why they are called general purpose.
This behaviour is perfectly analogous with how scripts work: Read and execute access ->./script.pl just read access -> perl script.pl
The goal of security work should not be to limit what can be executed, but who can execute. If random kids can execute arbitrary commands through your firefox, they can do whatever you can. Limiting their power in that scenario would also be limiting your own power in the same way.
I don't see a lot of people griping when their Closed-Source ATI linux driver keeps their video card running
I always make very sure my hardware works with Free drivers. Partly because of ideology, and partly of practical reasons. I know that I am not alone in this, I think most people who actually care about Free software make similar purchasing decisions. The reason you don't see us whining about it is that it is now quite possible to build and upgrade systems running 100% Free code.
Their selling point is integration. I have a client who design, sell and admin content management systems, and they are a pure MS shop. Their products rely heavily on the integration between IIS, Exchange,.Net and Active Directory. From what I have seen they would have a lot less hassle with a system of separate components that actually work and fit with their product instead of shoehorning their stuff into the MS conventions, but they are convinced that the superior integration of Microsoft's offerings give them a significant advantage.
On the other hand, they have pretty good knowledge about how to program for Microsoft products, and no knowledge whatsoever about any competing products, so in their case they are obviously better off using the stuff they know.
That's interesting. I used to do image retouching of my neg scans with Photoshop 6 on a 1.8GHz/512MB machine at work, and it was terribly slow. Then I discovered that doing it with the Gimp on my 700MHz/768MB machine at home was quite snappy. It was easily an order of magnitude faster, and I could keep several images open at once, something that would totally stop the Photoshop machine. Granted, these files weren't 1GB, but not exactly small either at around 200MB each.
I guess it just goes to show that YMMV.
(Disclaimer: The photoshop machine was running WinXP Pro, and the Gimp machine Linux 2.4, but that can't realistically account for the huge difference in speed I experienced)
And, yes, I do know enough fourier analysis to be aware of the Nyqvist frequency, but I don't know enough psychoacoustics to be sure that frequencies above 22kHz don't affect our listening experience, there are some who claim they do, and some who claim they don't.
In any case, it's pretty moot, since most cd's sound bad simply because the recording has been poorly mixed. I have a few very well-produced cd's and they sound very good to my ears.
And by "full lossless audio" you mean CD's, I presume? Hardly lossless, I'd say. And despite all the hype about FLAC and similar formats, I am pretty sure a high-quality frequency based format (as opposed to sample based PCM. MP3, Vorbis and ATRAC are all frequency based) could offer much better audio quality per data rate. Providing of course you don't first pass the recording through another lossy scheme such as 44.1kHz/16bit PCM.
Not that most of us have equipment capable of reproducing much higher quality than what fits on a CD, but still.
I couldn't resist firing up a vmware instance of Windows and try it with Internet Explorer. I even took a snapshot of the image before entering the address, real paranoid like.
Nothing. It displayed a blank page which reloaded itself at a moderate rate. I clicked stop and it stopped.
My brother and I used to compete in good mileage. Same car (a SAAB 900) and the same route from our parents' house in the outer suburbs to his apartment in central Stockholm, and we'd let the onboard computer calculate our average petrol usage. By adjusting our driving we managed to get it down to 8.2 l/100km, compared to the 9-ish we got in our first try.
Techniques included releasing the gas entirely in downhills (but keeping it in gear, this way virtually no fuel is used, the momentum keeps the engine going), accelerating ahead of upward slopes and then drop back down to cruising speed during the slope, carefully timed approaches to stop lights and other traffic. You get the idea, plan ahead and drive smooth. Top cruising speed made less of a difference.
All in all, this driving style resulted in a smooth and comfortable ride with good mileage and good safety margins. It also got us the fastest travel times.
That's pretty much what I meant. I've done some network security work for my cousin who runs a small software firm, and that's just fine. I had the feeling the poster was asking about helping friends out with their home computers. I can't justify charging $200 to fix someone's home computer, it isn't worth it, when the local shop will do it for half. Pretty much; don't do work you're overqualified for, unless it's as a favour. If you don't want to do favours, then don't. Doing half-favours for less than your going rate is not good for your friendship, and overcharging your friends for services they don't need is just plain mean.
My advice is (and this is much more general than just tech support), don't try and find a middle ground between business and family/friendship. Either you help them out because you want to, and don't charge for it (but make sure they remember it when you need help with something:), or you do it for the money, and charge what you feel it's worth. What do you say when you've charged money for helping your uncle with his computer, and you're unable to fix it? A normal customer can just say "screw you" and go elsewhere, but with family it isn't that simple.
I do some tech consulting (professionally) on the side, and usually charge my clients the equivalent of US$100-150 per hour, depending on the issue. Now, a hundred bucks an hour would be outrageous for some simple home PC fiddling, and I can't charge my friends that kind of money. This means I can't get enough money out of it to make it worth my time, so I would still be doing it as a favour. As long as I'm doing favours, I don't feel right charging for them.
First off, storage chip capacity is always measured in bits, and these chips are most likely (no, I haven't read the FA) 8Gib (eight gibibit, or 8x2^30 bits, i.e. 2^30 bytes or 1GiB). Look at your memory sticks and you will notice that they are equipped with mutliple memory chips, and the same goes for large capacity flash cards. This is not a 1GiB storage card, but a component that can be used to build large storage cards.
What is new and interesting (for chip process nerds) about this is the 60nm process. Current chips are generally produced in a 130nm process. There are also 90nm chips in general production, and they're pretty much the shitnitz at the moment.
I suppose it differs then. I can of course only speak of the people I know. Out of nine women I know who have given birth the last few years, only one took more than a month off before birth that I know of. Two of them took less than a week off (but they're workaholics both of them, staying home with their kids is going to do them a world of good).
Perhaps they should have taken more time off, health-wise, but all of them wanted to continue with their lives for as long as possible. The three days of your grandmother's must have been tough, though.
Oh, and I'm from Sweden, there might just be a cultural difference.
Of course the mother is out of work in the last months before the pregnancy (I'll assume you actually meant last months of her pregnancy)
No, she isn't. Pregnancy is not a disease. Depending of course what kind of a job you have, you won't need all that much time off before giving birth (If you're a professional drunkard, or shoot cannons at your stomach at carnivals, it's a different thing).
All new fathers I know have taken two to four weeks off in connexion with their childs' births, which would likely match the time the mother is away during her pregnancy. After that, as you point out, most people split the year between them. Mostly because after six months, the mother wants to get back to her job.
Slashdot Mirror
Yes, the unit is the same (obviously, as they are both expressions for energy), but I interpreted the grandparent (or whatever it is) as drawing more far-reaching conclusions from the similarity of the two expressions. Rereading it now I'm not sure what he was trying to say.
SUV's are usually much more rigid, so while most modern cars are designed to crumple in a way as to minimize the decelerating force on the passengers, the entire SUV and its occupants will decelerate at essentially the same rate. While someone travelling in an ordinary car is 3-4 times more likely to die from crashing into a SUV than from crashing into another car, the SUV driver is also at a greater risk than a car driver. This is mostly because SUV's often have poor safety systems, eg no whiplash protection, bad handling, no anti-skid systems and so on.
A majority of all traffic deaths are from single vehicle accidents, and here the SUV's are much more dangerous than ordinary cars. Again because of rigidity, but also as an effect of their greater mass.
So, to sum it up, the SUV drivers are the much safer party in SUV-car collisions, as they increase the risk of other drivers more than the risk to themselves, but all in all run a greater risk of dying in a car wreck.
(The numbers in this post come from the Swedish insurance agency Folksam, http://www.folksam.se/. Not a promotion, simply a source reference)
"This incidentally does make it in the same class as E=m*c^2 (due to magnitude/exponentiation) [...]"
No, it doesn't work that way, as c is a constant. E=mc^2 scales linearly with mass and does not include speed, as it deals with the energy of mass at rest. E=½mv^2 scales linearly with mass and squarely with speed.
(What is the rationale behind slashcode removing ² - or ² - from posts?)
I should have used preview, slashdot mangled that a bit. Second try.
Letting v=1 does not make gamma=1:
gamma = ( sqrt( 1 - (v/c)^2 ) )^-1
so:
gamma=1 <=> v=0
Which means that the kinetic energy is zero when the velocity is zero.
Letting v=1 does not make gamma=1:
gamma = ( sqrt( 1 - (v/c)^2 ) )^-1
so:
gamma=1 v=0
Which means that the kinetic energy is zero when the velocity is zero.
I have noticed that a couple of my users thouroughly enjoy the sight of GL screensavers (using xscreensaver), and they have never reported any lockups. I remind them regularly to report any and all annoyances with their computers, and they're usually pretty good at dong so, so I am fairly confident that there has been no lockups. I am currently running Xorg 6.8.2.
Perhaps you should try a driver upgrade?
I admin a score of machines using Sapphire 9200 cards, all running with the Xorg driver. The machines are used daily (I am posting this from one of them) and I have yet to see a single problem with the driver. Granted, OpenGL is mostly used for screen savers on these boxes, but still. In my experience the drivers are rock solid.
Well, it wasn't entirely meaningless, as you understood exactly what I meant :)
I didn't intend it as a great revelation to solve all our security woes, but merely as a reminder of what the original problem we are trying to solve is. One should always keep the original objective in mind when attacking sub-problems. It is all too easy to get lost in the details and go against it.
How true.
/mnt/unexecutables/evil.pl) example)
(Also, I must admit I don't understand your perl (cat
One has to remember that a program is just a sequence of instructions written down in a convenient form. Someone with access to a general purpose system can feed it general instructions. If your particular set of instructions are readable to him (which has been assumed so far in this discussion), he can feed them to the computer. He can do so with instructions he read on the Net, or on the back of a milk carton.
Perhaps what you are suggesting is that perl be configured so that it will only run when invoked from the script interpretor (i.e. when it is mentioned in a #!-statement in an executable text file). This would perhaps be an extra hoop to jump through, but would also greatly limit the usefullness of perl. I use perl -e 'something' all the time.
It probably would. Ponder the following:or
I don't consider this a design flaw, quite the opposite. Any general purpose operating system can always be told to execute arbitrary commands, that is why they are called general purpose.
./script.pl
This behaviour is perfectly analogous with how scripts work:
Read and execute access ->
just read access -> perl script.pl
The goal of security work should not be to limit what can be executed, but who can execute. If random kids can execute arbitrary commands through your firefox, they can do whatever you can. Limiting their power in that scenario would also be limiting your own power in the same way.
I always make very sure my hardware works with Free drivers. Partly because of ideology, and partly of practical reasons. I know that I am not alone in this, I think most people who actually care about Free software make similar purchasing decisions. The reason you don't see us whining about it is that it is now quite possible to build and upgrade systems running 100% Free code.
Their selling point is integration. I have a client who design, sell and admin content management systems, and they are a pure MS shop. Their products rely heavily on the integration between IIS, Exchange, .Net and Active Directory. From what I have seen they would have a lot less hassle with a system of separate components that actually work and fit with their product instead of shoehorning their stuff into the MS conventions, but they are convinced that the superior integration of Microsoft's offerings give them a significant advantage.
On the other hand, they have pretty good knowledge about how to program for Microsoft products, and no knowledge whatsoever about any competing products, so in their case they are obviously better off using the stuff they know.
That's interesting. I used to do image retouching of my neg scans with Photoshop 6 on a 1.8GHz/512MB machine at work, and it was terribly slow. Then I discovered that doing it with the Gimp on my 700MHz/768MB machine at home was quite snappy. It was easily an order of magnitude faster, and I could keep several images open at once, something that would totally stop the Photoshop machine. Granted, these files weren't 1GB, but not exactly small either at around 200MB each.
I guess it just goes to show that YMMV.
(Disclaimer: The photoshop machine was running WinXP Pro, and the Gimp machine Linux 2.4, but that can't realistically account for the huge difference in speed I experienced)
Yes, that was exactly my point :)
And, yes, I do know enough fourier analysis to be aware of the Nyqvist frequency, but I don't know enough psychoacoustics to be sure that frequencies above 22kHz don't affect our listening experience, there are some who claim they do, and some who claim they don't.
In any case, it's pretty moot, since most cd's sound bad simply because the recording has been poorly mixed. I have a few very well-produced cd's and they sound very good to my ears.
And by "full lossless audio" you mean CD's, I presume? Hardly lossless, I'd say. And despite all the hype about FLAC and similar formats, I am pretty sure a high-quality frequency based format (as opposed to sample based PCM. MP3, Vorbis and ATRAC are all frequency based) could offer much better audio quality per data rate. Providing of course you don't first pass the recording through another lossy scheme such as 44.1kHz/16bit PCM.
Not that most of us have equipment capable of reproducing much higher quality than what fits on a CD, but still.
I couldn't resist firing up a vmware instance of Windows and try it with Internet Explorer. I even took a snapshot of the image before entering the address, real paranoid like.
Nothing. It displayed a blank page which reloaded itself at a moderate rate. I clicked stop and it stopped.
My brother and I used to compete in good mileage. Same car (a SAAB 900) and the same route from our parents' house in the outer suburbs to his apartment in central Stockholm, and we'd let the onboard computer calculate our average petrol usage. By adjusting our driving we managed to get it down to 8.2 l/100km, compared to the 9-ish we got in our first try.
Techniques included releasing the gas entirely in downhills (but keeping it in gear, this way virtually no fuel is used, the momentum keeps the engine going), accelerating ahead of upward slopes and then drop back down to cruising speed during the slope, carefully timed approaches to stop lights and other traffic. You get the idea, plan ahead and drive smooth. Top cruising speed made less of a difference.
All in all, this driving style resulted in a smooth and comfortable ride with good mileage and good safety margins. It also got us the fastest travel times.
That's pretty much what I meant. I've done some network security work for my cousin who runs a small software firm, and that's just fine. I had the feeling the poster was asking about helping friends out with their home computers. I can't justify charging $200 to fix someone's home computer, it isn't worth it, when the local shop will do it for half. Pretty much; don't do work you're overqualified for, unless it's as a favour. If you don't want to do favours, then don't. Doing half-favours for less than your going rate is not good for your friendship, and overcharging your friends for services they don't need is just plain mean.
My advice is (and this is much more general than just tech support), don't try and find a middle ground between business and family/friendship. Either you help them out because you want to, and don't charge for it (but make sure they remember it when you need help with something :), or you do it for the money, and charge what you feel it's worth. What do you say when you've charged money for helping your uncle with his computer, and you're unable to fix it? A normal customer can just say "screw you" and go elsewhere, but with family it isn't that simple.
I do some tech consulting (professionally) on the side, and usually charge my clients the equivalent of US$100-150 per hour, depending on the issue. Now, a hundred bucks an hour would be outrageous for some simple home PC fiddling, and I can't charge my friends that kind of money. This means I can't get enough money out of it to make it worth my time, so I would still be doing it as a favour. As long as I'm doing favours, I don't feel right charging for them.
"The way I see it I'm selling something I already own"
Wow, that's neat. I generally am only able to sell things I not yet own. But of course, once I've sold it, it's mine!
(And yes, I turned off my karma bonus for this silly posting)
First off, storage chip capacity is always measured in bits, and these chips are most likely (no, I haven't read the FA) 8Gib (eight gibibit, or 8x2^30 bits, i.e. 2^30 bytes or 1GiB). Look at your memory sticks and you will notice that they are equipped with mutliple memory chips, and the same goes for large capacity flash cards. This is not a 1GiB storage card, but a component that can be used to build large storage cards.
What is new and interesting (for chip process nerds) about this is the 60nm process. Current chips are generally produced in a 130nm process. There are also 90nm chips in general production, and they're pretty much the shitnitz at the moment.
I suppose it differs then. I can of course only speak of the people I know. Out of nine women I know who have given birth the last few years, only one took more than a month off before birth that I know of. Two of them took less than a week off (but they're workaholics both of them, staying home with their kids is going to do them a world of good).
Perhaps they should have taken more time off, health-wise, but all of them wanted to continue with their lives for as long as possible. The three days of your grandmother's must have been tough, though.
Oh, and I'm from Sweden, there might just be a cultural difference.
Of course the mother is out of work in the last months before the pregnancy
(I'll assume you actually meant last months of her pregnancy)
No, she isn't. Pregnancy is not a disease. Depending of course what kind of a job you have, you won't need all that much time off before giving birth (If you're a professional drunkard, or shoot cannons at your stomach at carnivals, it's a different thing).
All new fathers I know have taken two to four weeks off in connexion with their childs' births, which would likely match the time the mother is away during her pregnancy. After that, as you point out, most people split the year between them. Mostly because after six months, the mother wants to get back to her job.