it is quite trivial to turn a biased rng into an unbiased one.
1) generate pairs of 1's,0's (assume 80% are ones) (assume there is no correlation between the two numbers generated)
2) within the number stream every time 01 occurs, output 0, every time 10 occurs output 1, 11 and 00 output nothing
01 occurs.2*.8 (16%) of the time, 10 occurs.8*.2 (16%) of the time, 11 occurs 64% of the time, 00 occurs 4% of the time
therefore the number of 1's and 0's output by step 2 is the same, at the cost of 84% of the generating capacity (with an unbiased generator all 4 of the numbers above are 25%, and this scheme results in the loss of 75% of generating capacity)
1) She wasn't selling. (as far as i know)
2) In this digital age, there are no copies actually made until someone actually downloaded.
I don't have a problem with the guilty verdict either, but having to pay 5 years income (based on an average income of w/o $45k) w/o any actual damages having to be demonstrated seems a little unbalanced.
just for interest, how do you seed gentoo, beyond the first stage 3 install? - if you are talking about a pre-installed gentoo based live dvd, such as sabayon, or hosting a portage mirror, i could understand. I'm just not sure how you seed gentoo.
but it was obviosly worth both the electrical power for the compuuter for 40 seconds, and the rather more significant (compared to the neural activity) electro-chemical expenditure of typing that response. This typifies current american philosophy in which it seems to be better to exert effort to be seen to not be thinking, rather than to just do the thinking in the first place.
i somehow suspect that any sensabil filter has slashdot on the "omg, not again, ignore it" list, but then again a) look how many people are here, statistally someone here is up to no good, and b) this is the gov. we're talking about....
and i'm sure that car 'theft' would similarly be quite common if there was a way to duplicate a car with no inconvenience to the person who owned the original, for less than the cost of a can of coke.
i beg to differ - in may cases ray tracing is an embarisngly parallel problem - one ray for each pixel of the screen, completely independently of the others (ignoring AA type effects w/ multiple ray per screen pixel)
you just have to have a good set of shared memory behind it, as the memory access is random across a large area with this scheme
That's not completely true, there are a few (minor) areas where the windows kernel has features that linux doesn't.
An objective (feature by feature) comparison can be found at http://widefox.pbwiki.com/Kernel%20Comparison%20Li nux%20vs%20Windows.
The current comparison is Vista vs 2.6.22 (features that can be patched in are noted), and it looks like the next update will be Linux 2.6.23 kernel versus Windows Server 2008 kernel.
congratulations to #292 for winning todays Low ID contest
(at least my ID isn't 6 digit)
--
The early bird gets the worm, but the second mouse gets the cheese
yes, but you can use a few >90nm components to completely shut off power to specific cores to result in no leakage current, at the cost of a few clock cycles initialization time when you power it up again.
I never said "routers are devices that hand out IP's", i just noted that it is a common feature. I don't know of any (currently sold) router which does not have a DHCP server built in. (although the routers you buy at bestbuy would probably be the place to find such a thing.)
I intended my post to imply that the primary location for DHCP servers was within devices acting as routers, with a few also being located in non-routing (file/domain/...) servers.
even my consumer grade dlink adsl-router runs DHCP, which allows it to assign anyone connecting to it an IP address.
Given that network equipment consists primarily of servers, workstations, routers, switches and hubs, and that of those only servers and routers are likely to be able to give out ip addresses, i would guess that most IP's in the world were actually given out by routers.
The question remains why run both windows and linux?
I run windows for ease of game playing (yes i know most things can be convinced to work under linux, but it is less time to maintain a windows install than it is to get them working).
The biggest problem with virtualisation technology is that there is minimal 3d acceleration within the guest. This leads me to run windows native, for the game performance when needed, and have linux running in a virtual machine. As linux can also be compiled paravertualised (in the newer kernels,.21 and newer i think) this leads to linux-in-windows being more efficent in many situations than windows-in-linux, both from the functionality-of-guest, and game-playing points of view.
If anyone knows otherwise, please let me know.
Andy
I would guess that the communication overhead would be quite dependent on design, but would be relatively large. This is the reason that it is 1 monolithic program, but my post was more concerned with it being a theoretical possibility, as compared to the great-GP who (in a somewhat snarky manner) used FF as an example of a program that *could not* be broken up.
I wonder if it would be possible to design an inter-process communication scheme in which it would be practical to break a large program like FF down into 10 or so sub-threads?
- GUI stuff - network communication stuff - parsing (and running, if needed) stuff (xml/javascript/dhtml/...)
theres three relatively discrete components, which are (theoretically anyway) only connected via requests for data and replies - i'm sure that each of those could be broken down further (plausible independent subsets of 'GUI stuff' could include menus, action bars, the actual webpage)
"person x does something, person y find out about it and an action results". is this representative of the problem statement? or is it "person x does something, person z has a record of the event; possibly publically avaible."
if the only objection to the second statement is that it modifies the ease at which the first can result, then you have a problem with the first statemnt. To resolve that issue, simply don't behave in an indecent manner (in the above example the embarising action is "getting drunk and looking like a fool"), or perform any action with other people which you would not want potentially everyone to know about.
The default state for knowlege is to spread - the entire idea that you can place artifical boundries on the spread of information is what thw whole DRM mess is about.
If other people have photos of you you don't want shared, just ask them - most people are honerable enough to not do so, and if not then why would you perform such actions infront of them?
yeah, the problems people have with other people trying to give you the video w/o actually giving you the video. (ps, does anyone have a tool to get a video wrapped in a swf file?)
actually, that's a bit extreme, all you need to do is to heat it above the curie temperature (300-380 for Fe-Nd alloys) at this point the magnetic properties become completely dependent on the applied magnetic field, so as it cools down again, the only magnetization left is due to the earths magnetic field.
Below this temperature you need to apply a strong magnetic field to reverse *most* of the magnetization (thats how normal recording works).
As an added bonus if you do this in such a way as there are not dust contaminants (inductive heating of the platters in a vacuum) you still have a working drive.
Re:Exactly...it proves nothing....
on
A Mighty Number Falls
·
· Score: 2, Interesting
actually, although quantum computers can theoretically factor huge numbers trivially, quantum computers are _extremely_ sensitive to error, and as such the result is statistical, with the chance of a right answer decreasing with the size of the problem.
To factor a RSA prime would need about a 1 million qbit computer (the extra bits are needed for error correction), and at the moment a 10 qbit quantum computer hasn't been constructed, and current design principles don't scale well beyond about 1k qbits. as the hardness of building a quantum computer is proportional to the number of qbits it has to process (due to having to shield it from all outside interference) it is still quite feasible to increase the size of the prime to a length where it is still 'hard' to break, although increases in technology would mearly give you years of security, rather than the "practically forever" security that RSA currently has.
Slashdot Mirror
it is quite trivial to turn a biased rng into an unbiased one. 1) generate pairs of 1's,0's (assume 80% are ones) (assume there is no correlation between the two numbers generated) 2) within the number stream every time 01 occurs, output 0, every time 10 occurs output 1, 11 and 00 output nothing 01 occurs .2*.8 (16%) of the time, 10 occurs .8*.2 (16%) of the time, 11 occurs 64% of the time, 00 occurs 4% of the time
therefore the number of 1's and 0's output by step 2 is the same, at the cost of 84% of the generating capacity (with an unbiased generator all 4 of the numbers above are 25%, and this scheme results in the loss of 75% of generating capacity)
1) She wasn't selling. (as far as i know) 2) In this digital age, there are no copies actually made until someone actually downloaded. I don't have a problem with the guilty verdict either, but having to pay 5 years income (based on an average income of w/o $45k) w/o any actual damages having to be demonstrated seems a little unbalanced.
just for interest, how do you seed gentoo, beyond the first stage 3 install? - if you are talking about a pre-installed gentoo based live dvd, such as sabayon, or hosting a portage mirror, i could understand. I'm just not sure how you seed gentoo.
but it was obviosly worth both the electrical power for the compuuter for 40 seconds, and the rather more significant (compared to the neural activity) electro-chemical expenditure of typing that response. This typifies current american philosophy in which it seems to be better to exert effort to be seen to not be thinking, rather than to just do the thinking in the first place.
i somehow suspect that any sensabil filter has slashdot on the "omg, not again, ignore it" list, but then again a) look how many people are here, statistally someone here is up to no good, and b) this is the gov. we're talking about ....
and i'm sure that car 'theft' would similarly be quite common if there was a way to duplicate a car with no inconvenience to the person who owned the original, for less than the cost of a can of coke.
not to mention the fun and games it would take to keep the side in shadow from freezing solid, and the side in the sun from boiling
i beg to differ - in may cases ray tracing is an embarisngly parallel problem - one ray for each pixel of the screen, completely independently of the others (ignoring AA type effects w/ multiple ray per screen pixel) you just have to have a good set of shared memory behind it, as the memory access is random across a large area with this scheme
That's not completely true, there are a few (minor) areas where the windows kernel has features that linux doesn't.i nux%20vs%20Windows.
An objective (feature by feature) comparison can be found at http://widefox.pbwiki.com/Kernel%20Comparison%20L
The current comparison is Vista vs 2.6.22 (features that can be patched in are noted), and it looks like the next update will be Linux 2.6.23 kernel versus Windows Server 2008 kernel.
"This should make it much easier for NASA to lose priceless historical data ... "
you mean items like the original Apollo 13 moonwalk tapes?
congratulations to #292 for winning todays Low ID contest (at least my ID isn't 6 digit) -- The early bird gets the worm, but the second mouse gets the cheese
yes, but you can use a few >90nm components to completely shut off power to specific cores to result in no leakage current, at the cost of a few clock cycles initialization time when you power it up again.
I never said "routers are devices that hand out IP's", i just noted that it is a common feature. I don't know of any (currently sold) router which does not have a DHCP server built in. (although the routers you buy at bestbuy would probably be the place to find such a thing.) I intended my post to imply that the primary location for DHCP servers was within devices acting as routers, with a few also being located in non-routing (file/domain/...) servers.
even my consumer grade dlink adsl-router runs DHCP, which allows it to assign anyone connecting to it an IP address. Given that network equipment consists primarily of servers, workstations, routers, switches and hubs, and that of those only servers and routers are likely to be able to give out ip addresses, i would guess that most IP's in the world were actually given out by routers.
still no salt in that particular wound? - i would of thought that they would of learned from their mistakes, at least where it is easy to fix.
Try IE tab - it allows you to switch rendering engines of a specific FF tab/window1 9
https://addons.mozilla.org/en-US/firefox/addon/14
The question remains why run both windows and linux? I run windows for ease of game playing (yes i know most things can be convinced to work under linux, but it is less time to maintain a windows install than it is to get them working). The biggest problem with virtualisation technology is that there is minimal 3d acceleration within the guest. This leads me to run windows native, for the game performance when needed, and have linux running in a virtual machine. As linux can also be compiled paravertualised (in the newer kernels, .21 and newer i think) this leads to linux-in-windows being more efficent in many situations than windows-in-linux, both from the functionality-of-guest, and game-playing points of view.
If anyone knows otherwise, please let me know.
Andy
have you tried Ubuntu: The Worlds Best point-and-click adventure game?
I would guess that the communication overhead would be quite dependent on design, but would be relatively large. This is the reason that it is 1 monolithic program, but my post was more concerned with it being a theoretical possibility, as compared to the great-GP who (in a somewhat snarky manner) used FF as an example of a program that *could not* be broken up.
I wonder if it would be possible to design an inter-process communication scheme in which it would be practical to break a large program like FF down into 10 or so sub-threads?
- GUI stuff
- network communication stuff
- parsing (and running, if needed) stuff (xml/javascript/dhtml/...)
theres three relatively discrete components, which are (theoretically anyway) only connected via requests for data and replies - i'm sure that each of those could be broken down further (plausible independent subsets of 'GUI stuff' could include menus, action bars, the actual webpage)
"Upon completion of this investigation, Microsoft ... may [issue] a security advisory"
"person x does something, person y find out about it and an action results".
is this representative of the problem statement?
or is it
"person x does something, person z has a record of the event; possibly publically avaible."
if the only objection to the second statement is that it modifies the ease at which the first can result, then you have a problem with the first statemnt.
To resolve that issue, simply don't behave in an indecent manner (in the above example the embarising action is "getting drunk and looking like a fool"), or perform any action with other people which you would not want potentially everyone to know about.
The default state for knowlege is to spread - the entire idea that you can place artifical boundries on the spread of information is what thw whole DRM mess is about.
If other people have photos of you you don't want shared, just ask them - most people are honerable enough to not do so, and if not then why would you perform such actions infront of them?
yeah, the problems people have with other people trying to give you the video w/o actually giving you the video. (ps, does anyone have a tool to get a video wrapped in a swf file?)
actually, that's a bit extreme, all you need to do is to heat it above the curie temperature (300-380 for Fe-Nd alloys) at this point the magnetic properties become completely dependent on the applied magnetic field, so as it cools down again, the only magnetization left is due to the earths magnetic field. Below this temperature you need to apply a strong magnetic field to reverse *most* of the magnetization (thats how normal recording works). As an added bonus if you do this in such a way as there are not dust contaminants (inductive heating of the platters in a vacuum) you still have a working drive.
actually, although quantum computers can theoretically factor huge numbers trivially, quantum computers are _extremely_ sensitive to error, and as such the result is statistical, with the chance of a right answer decreasing with the size of the problem. To factor a RSA prime would need about a 1 million qbit computer (the extra bits are needed for error correction), and at the moment a 10 qbit quantum computer hasn't been constructed, and current design principles don't scale well beyond about 1k qbits. as the hardness of building a quantum computer is proportional to the number of qbits it has to process (due to having to shield it from all outside interference) it is still quite feasible to increase the size of the prime to a length where it is still 'hard' to break, although increases in technology would mearly give you years of security, rather than the "practically forever" security that RSA currently has.