Here's the difference between a test that shows correlation and causation:
In the test that is able to show causation the experimenter got to select the experimental and control groups (in a statistically appropriate way) and then was able to make sure that the two groups only varied as called for by the experiment.
In order to show a causual link between violent video games and violent behaviour, you'd need to select your groups and declare a control group and an experimental group - not just pick out a group of people who look a bit like your experiment after the fact ("sure, those people played video games and... these people didn't")
One of the basic properties that defines a free market economy is the lack of government-granted monopolies. By that guideline the telecom industry completely fails at being even vaguely related to a free market.
Screwing up randomization on a computer is really hard. Algorithims for correctly reandomizing a list are very well known, and any random programmer should be able to come up with them without any references.
The only potentially hard part is generating random numbers, but seeded computer psudorandom number generators are more than good enough that - given a changing seed (even the time from the system clock) - they will produce a random looking result that is more than sufficient for randomizing a music playlist or any other application where you don't need to worry about cryptanalysis.
It's probably possible to play the single player campaigns with just a mouse, especially if you turn down the speed, but StarCraft is really a multiplayer game and you need a keyboard for that.
Bah! 56K! I used to dial into my friend's computer using my 28.8 modem to play Descent. Whippersnapper./me waits for 5200 baud fogies with double digit UIDs to crawl from the woodwork
5200 baud? I remember when we had to transmit our data by morse code *and* spell out the zeroes and ones.
dawdawditdit dit ditdawdit dawdawdaw dawdawdaw dawdit dit dawdawdaw dawdit dit dawdawditdit dit ditdawdit dawdawdaw dawdawditdit dit ditdawdit dawdawdaw dawdawditdit ditditdawdit dawdawdaw dawdawditdit dit ditdawdit dawdawdaw dawdawdaw dawdit dit
Was just the letter 'a'!
On the other hand, our operators could go almost 100 WPM. I guess that comes out to a total of about 1 and a third baud.
And now I have this new-fangled message board system telling me that morse code is junk. Obviously some ill-educated ruffian programmed it.
When I was a kid, I had to walk to school in the snow barefoot uphill both ways forty miles.
At least we didn't have these horrible asymetric services you kids pay for these days. The operator on the other end could send back a response just as fast as it came in.
I don't care what the "horrible disaster of the day" is, it doesn't make spending money on anything but charity an evil selfish act.
Apparently some people believe that Enterprise is a good enough TV show that it is worth their money to try to save it. I'm not sure that a newspaper ad is the correct way to spend that money, but supporting something you enjoy is a perfectly reasonable and appropriate thing to do.
In some alternate universe where the only use for accelerated 3d was gaming, your post would make tons of sense.
Here in the real world, hardware accelerated 3d is an important capibility for everything from CAD to basic 2d desktop rendering.
The requirement for 3d hardware acceleration for general usage applications is becoming more and more widespread. Already features that were only avaiable in high-end 3d cards in 1995 are now required to get a reasonable user experiance out of both Windows XP and Mac OS X - I wouldn't assume that modern Open Source desktop environments won't use the same techniques to keep up.
Since they "fixed" the sid=anything "bug", this is a bit harder. Before, an amusing extention to the bittorrent protocal would have been to post the torrent file contents as a comment under sid=hash, and then if the tracker went down to post a new comment under the same sid with the new tracker.
I don't know about you, but if I'm spending 10's of thousands of dollars for a software licence, I'll go through the extra 6 or 8 hours of obnoxiousness to get RHEL or SUSE up and working if that's what's officially supported.
I prefer Debian. Significantly. But for $30,000+, milking every penny out of Oracle's tech support is more important than my distro preference.
You can get a good keyboard for just a little over $5 now, but a good mouse is still in the $20 range. If you want a mouse that actually has near screen resolution (i.e. doesn't jump pixels), you'll need to pay a bit more - probably around $40.
It's a legitimate problem for powerbooks, where the built in mouse only has one button.
"Just connect an external mouse!" you say? Well, try using an external mouse while sitting crosslegged on a floor. Doesn't work for me - hence they've lost a sale of their product over the lack of a two or three button mouse.
Remember: Heat dissipation (aka. Power Consumption) is directly relevent to building a system. I don't know how you'd cool an 8-core Pentium IV (What, 600 Watts or something?), on the other hand - an 8-core Pentium M might only take 200 Watts - similar to a high-end graphics card. And if you could get 600 Watts, why would the Pentium IV * 8 be better than the Pentium M * 24 (or the Pentium M LV * 60).... This is all assuming your application is parallelizable, but most of the raw number crunching that a Pentium IV is good at is pretty easily paralleized.
Slashdot Mirror
Wrong.
Here's the difference between a test that shows correlation and causation:
In the test that is able to show causation the experimenter got to select the experimental and control groups (in a statistically appropriate way) and then was able to make sure that the two groups only varied as called for by the experiment.
In order to show a causual link between violent video games and violent behaviour, you'd need to select your groups and declare a control group and an experimental group - not just pick out a group of people who look a bit like your experiment after the fact ("sure, those people played video games and... these people didn't")
One of the basic properties that defines a free market economy is the lack of government-granted monopolies. By that guideline the telecom industry completely fails at being even vaguely related to a free market.
Define "cheap" as "sub-$250" instead, and then you'll see things the way I do.
Screwing up randomization on a computer is really hard. Algorithims for correctly reandomizing a list are very well known, and any random programmer should be able to come up with them without any references.
The only potentially hard part is generating random numbers, but seeded computer psudorandom number generators are more than good enough that - given a changing seed (even the time from the system clock) - they will produce a random looking result that is more than sufficient for randomizing a music playlist or any other application where you don't need to worry about cryptanalysis.
Actually, you could easily solve the halting problem for computers with a metacomputer.
Now all you need to do is make that last sentance mean something.
Don't complain. It's the people making that call who are driving the price decrease that lets you buy the card cheap 6 months later.
It's probably possible to play the single player campaigns with just a mouse, especially if you turn down the speed, but StarCraft is really a multiplayer game and you need a keyboard for that.
5200 baud? I remember when we had to transmit our data by morse code *and* spell out the zeroes and ones.
dawdawditdit dit ditdawdit dawdawdaw dawdawdaw dawdit dit dawdawdaw dawdit dit dawdawditdit dit ditdawdit dawdawdaw dawdawditdit dit ditdawdit dawdawdaw dawdawditdit ditditdawdit dawdawdaw dawdawditdit dit ditdawdit dawdawdaw dawdawdaw dawdit dit
Was just the letter 'a'!
On the other hand, our operators could go almost 100 WPM. I guess that comes out to a total of about 1 and a third baud.
And now I have this new-fangled message board system telling me that morse code is junk. Obviously some ill-educated ruffian programmed it.
When I was a kid, I had to walk to school in the snow barefoot uphill both ways forty miles.
At least we didn't have these horrible asymetric services you kids pay for these days. The operator on the other end could send back a response just as fast as it came in.
Just under 417 Kessel runs.
A USB key is probably pretty cheap, so just make the user pay to replace it. Simple enough.
That's just a normal episode plot. I wouldn't expect that to slow things down at all.
You sir are the sorry one.
I don't care what the "horrible disaster of the day" is, it doesn't make spending money on anything but charity an evil selfish act.
Apparently some people believe that Enterprise is a good enough TV show that it is worth their money to try to save it. I'm not sure that a newspaper ad is the correct way to spend that money, but supporting something you enjoy is a perfectly reasonable and appropriate thing to do.
Isn't JPEG better for that?
In some alternate universe where the only use for accelerated 3d was gaming, your post would make tons of sense.
Here in the real world, hardware accelerated 3d is an important capibility for everything from CAD to basic 2d desktop rendering.
The requirement for 3d hardware acceleration for general usage applications is becoming more and more widespread. Already features that were only avaiable in high-end 3d cards in 1995 are now required to get a reasonable user experiance out of both Windows XP and Mac OS X - I wouldn't assume that modern Open Source desktop environments won't use the same techniques to keep up.
Since they "fixed" the sid=anything "bug", this is a bit harder. Before, an amusing extention to the bittorrent protocal would have been to post the torrent file contents as a comment under sid=hash, and then if the tracker went down to post a new comment under the same sid with the new tracker.
One of. Try A instead of B if you don't want to deal with "any third party".
I don't know about you, but if I'm spending 10's of thousands of dollars for a software licence, I'll go through the extra 6 or 8 hours of obnoxiousness to get RHEL or SUSE up and working if that's what's officially supported.
I prefer Debian. Significantly. But for $30,000+, milking every penny out of Oracle's tech support is more important than my distro preference.
They turned that shit game into an ActiveX control? You can still download it though, so there's no need for ActiveX.
You have a single data point. With a couple more, you might even have a minor trend.
Hint: If Oracle isn't supported, there's no reason to be using it.
You can get a good keyboard for just a little over $5 now, but a good mouse is still in the $20 range. If you want a mouse that actually has near screen resolution (i.e. doesn't jump pixels), you'll need to pay a bit more - probably around $40.
It's a legitimate problem for powerbooks, where the built in mouse only has one button.
"Just connect an external mouse!" you say? Well, try using an external mouse while sitting crosslegged on a floor. Doesn't work for me - hence they've lost a sale of their product over the lack of a two or three button mouse.
Remember: Heat dissipation (aka. Power Consumption) is directly relevent to building a system. I don't know how you'd cool an 8-core Pentium IV (What, 600 Watts or something?), on the other hand - an 8-core Pentium M might only take 200 Watts - similar to a high-end graphics card. And if you could get 600 Watts, why would the Pentium IV * 8 be better than the Pentium M * 24 (or the Pentium M LV * 60). ... This is all assuming your application is parallelizable, but most of the raw number crunching that a Pentium IV is good at is pretty easily paralleized.
Less like freenet, more like bittorrent. There is no anonymity.
Arcfour is a legit algorithim for this, and it would work fine if they had gotten their cypher mode right.