How many people utilize their cycles beyond 10 or 20% over time?
Well, checking my system right now:
load average: 1.37, 1.46, 1.17
But that's bogus, anyway. Running a software DSP in drivers isn't a batch job that can be offloaded to idle time -- it's a constant drain. Not much sense in getting the latest and greatest CPU when you're throwing a significant chunk of CPU time out the window.
For what people save on their system buy going the winmodem route, they could just buy a six-month old system and a hardware modem and have some money left over and not deal with driver issues and get the same level of performance.
802.11b runs about 200 times as fast as a POTS modem. It also uses spread-spectrum. That's a lot more CPU time than Moore's law accounts for -- systems aren't even 200 times as fast as they were when winmodems were introduced.
Most winmodem drivers *are* horribly written. They're for budget buyers, and the manufacturer is not going to drop a bunch of money on driver dev *or* keep updating the driver. When we upgraded to win2k at work, one box with a winmodem just locked up periodically if we used the thing -- and no driver updates existed, at least six months after we upgraded. Sucks to lose all your work. Plus, the computer would temporarily freeze while dialing.
You don't usually see (especially not open source) Linux/other OS drivers for winmodems.
So if you go winmodem, you've just shackled yourself to the driver manufacturer on a low-budget item. Kind of like winprinters (but these at least now have open source Linux support more often, which means at least Linux users don't have to worry about losing the ability to use their printer in the future).
claiming that offloading the processing of an 802.11b card onto a 1.xGHz processor is going to "drag down" the system is a steaming pile
You do realize that there is a lot more signal analysis going on in an 802.11b (why do people call it "wi-fi"?) card than a winmodem -- it isn't just a direct linear function of the bandwidth, which is only about 200 times as much CPU time. 802.11b will be a *pain*. And the clean signal issue is just silly -- there's not a hell of a lot being generated on those cards, and in every system I know of there's so much electronic junk sitting around there already that the removal of a bit of circuitry isn't going to make much of a difference. If you were working with a system where the card was on a cable 6' from the computer you might have an argument.
I was just reading microwarehouse's catalog (which has fairly poor prices), and there were at least three external hardware modems under $100. One was $75.
Driver quality on cheapo winmodems (at least on Windows) is pretty piss-poor.
When we upgraded to 2k at work, six months later stable drivers *still* were't out for the damn winmodem in one box, and it occasionally froze. It also froze the entire computer while dialing.
If you buy an external modem, you know that you're getting solid, well tested hardware.
It cracks me up how people read benchmarks and try to get a system that runs 5% faster than another and then load the thing up with a winmodem, winprinter and Win XP and make the thing miserably slow again.
Actually, I don't use a winmodem, because they're crap hardware.
I have a nice external 3com modem. Actually, I *had* a nice external 3com modem until lightning hit, and now I have a nice external zoom modem.
I'm also glad that I had an external modem instead of something indirectly electrically connected to my PCI bus or I might be shelling out for a nice new computer right now.
I like the GPL, but it's a godawful license for taxpayer funded software. At the most, the software should be LGPL, and probably BSD or public domain would be best.
To do otherwise would screw over the legions of US closed source software companies. Open source should win fairly, through code quality, not because the gov't supports it with tons of money and makes it difficult to compete with closed source.
Um...unless Canadian criminal law differs significantly from US law, you are incorrect.
Firing a rifle at anyone is considered deadly force, period. Using a knife is considered deadly force. Furthermore, anything that can cause severe bodily harm, even if non-fatal, is considered deadly force in the United States. Raping someone is considered use of deadly force -- a potential rape victim is thus automatically allowed to use deadly force to avoid being raped.
There is no legal difference between "firing to wound" and "firing to kill".
Basically, in a self-defense situation, you are not allowed to use deadly force unless you have a reasonable belief (i.e. the jury feels that in your shoes they'd feel the same way) that you were about to be subjected to deadly force. In this case, you can use deadly force (such as a gun) to prevent deadly force from being used against you. You *cannot* use deadly force if the threat is not immediate -- if the person is running away from you to get away, you cannot shoot them in the back.
If nondeadly force is used against you, you may only use nondeadly force to prevent it. If someone says that they're going to give you a wedgie, you can shove them away and run or knock them to the ground to get away, but you cannot spray them with an automatic weapon.
There is one exception to weapons being considered deadly force -- *threatening* someone with a gun is not considered deadly force. You *may* threaten a burglar with a weapon if you catch him in your house. However, unless he reaches for a pocket (possibly a gun) or rushes you, you cannot actually legally fire the weapon.
(There is an exception to this in Texas, where you are allowed to use deadly force against burglars.)
This could differ in Canda, but Canadian law and US law are awfully similar.
I see your point, but lots of things directly do damage. The article author's point seemed to be that damage is the reason that virus writing/distribution is not okay.
So it comes down to whether we should ban something for practical reasons.
Guns and cars actually *kill* people, but they've been okayed under the don't-illegalize-it-unless-absolutely-necessary doctrine.
Frankly, the biggest malware threat I see is through worms spread by the technically illiterate -- in ICQ and email. I don't see redistribution of viruses as being all that big of a threat.
And I really dislike legislation against writing software, period. Make illegal the use of that software, okay, but not the writing of it. Code is just a little too close a the description of a thing. It feels like blueprints for bombs are being banned.
No one likes their damn accounts, no one likes being tracked, and there are a hell of a lot of other news sites. Please *stop* posting Slashdot articles with links to NYTimes articles. You just encourage them.
Seems that you stopped telemarketing as soon as you could.
The more miserable telemarketers there are the harder it is to get people to stay to do the job.
Besides, making telemarketers as unhappy as they made you is downright cathartic.
The "I'm just doing it to earn money" thing doesn't fly -- or else drugrunning, assassination, etc. would be kosher as well. Find a way to earn money that doesn't make people unhappy and they won't be mean back.
With a private institution, you *can* do exactly this.
The KKK is rather unlikely to accept a Black Panther as a member.
Frankly, I don't really have any problem at all with this. I don't even have a problem with discriminating against chunks of the population at public workplaces -- sorry, but if you're a guy, you just can't work at Hooters. That's just how it is. If Hooters is making a really stupid decision, then they'll go out of business eventually as their competitors kick their butt.
Slashdot Mirror
How many people utilize their cycles beyond 10 or 20% over time?
Well, checking my system right now:
load average: 1.37, 1.46, 1.17
But that's bogus, anyway. Running a software DSP in drivers isn't a batch job that can be offloaded to idle time -- it's a constant drain. Not much sense in getting the latest and greatest CPU when you're throwing a significant chunk of CPU time out the window.
For what people save on their system buy going the winmodem route, they could just buy a six-month old system and a hardware modem and have some money left over and not deal with driver issues and get the same level of performance.
802.11b runs about 200 times as fast as a POTS modem. It also uses spread-spectrum. That's a lot more CPU time than Moore's law accounts for -- systems aren't even 200 times as fast as they were when winmodems were introduced.
Most winmodem drivers *are* horribly written. They're for budget buyers, and the manufacturer is not going to drop a bunch of money on driver dev *or* keep updating the driver. When we upgraded to win2k at work, one box with a winmodem just locked up periodically if we used the thing -- and no driver updates existed, at least six months after we upgraded. Sucks to lose all your work. Plus, the computer would temporarily freeze while dialing.
You don't usually see (especially not open source) Linux/other OS drivers for winmodems.
So if you go winmodem, you've just shackled yourself to the driver manufacturer on a low-budget item. Kind of like winprinters (but these at least now have open source Linux support more often, which means at least Linux users don't have to worry about losing the ability to use their printer in the future).
Why do you care?
Microsoft is funding Slashdot to continue being anti-MS. Works for me.
claiming that offloading the processing of an 802.11b card onto a 1.xGHz processor is going to "drag down" the system is a steaming pile
You do realize that there is a lot more signal analysis going on in an 802.11b (why do people call it "wi-fi"?) card than a winmodem -- it isn't just a direct linear function of the bandwidth, which is only about 200 times as much CPU time. 802.11b will be a *pain*. And the clean signal issue is just silly -- there's not a hell of a lot being generated on those cards, and in every system I know of there's so much electronic junk sitting around there already that the removal of a bit of circuitry isn't going to make much of a difference. If you were working with a system where the card was on a cable 6' from the computer you might have an argument.
I was just reading microwarehouse's catalog (which has fairly poor prices), and there were at least three external hardware modems under $100. One was $75.
Driver quality on cheapo winmodems (at least on Windows) is pretty piss-poor.
When we upgraded to 2k at work, six months later stable drivers *still* were't out for the damn winmodem in one box, and it occasionally froze. It also froze the entire computer while dialing.
If you buy an external modem, you know that you're getting solid, well tested hardware.
It cracks me up how people read benchmarks and try to get a system that runs 5% faster than another and then load the thing up with a winmodem, winprinter and Win XP and make the thing miserably slow again.
Actually, I don't use a winmodem, because they're crap hardware.
I have a nice external 3com modem. Actually, I *had* a nice external 3com modem until lightning hit, and now I have a nice external zoom modem.
I'm also glad that I had an external modem instead of something indirectly electrically connected to my PCI bus or I might be shelling out for a nice new computer right now.
Yeah, well, you generally use a printer at the same time as another app, and always a soft modem.
I really loathe hardware-without-the-hardware.
I thought Arcanum and Fallout were both fun.
Never buy a game without playing the demo. If you like the demo, buy the game. Pretty strong correlation between the two.
If there's no demo, don't buy the game.
Do you know of any game review websites that have policies against accepting perks?
Hmm...that should be "difficult for closed source software to compete with it"
I like the GPL, but it's a godawful license for taxpayer funded software. At the most, the software should be LGPL, and probably BSD or public domain would be best.
To do otherwise would screw over the legions of US closed source software companies. Open source should win fairly, through code quality, not because the gov't supports it with tons of money and makes it difficult to compete with closed source.
Um...unless Canadian criminal law differs significantly from US law, you are incorrect.
Firing a rifle at anyone is considered deadly force, period. Using a knife is considered deadly force. Furthermore, anything that can cause severe bodily harm, even if non-fatal, is considered deadly force in the United States. Raping someone is considered use of deadly force -- a potential rape victim is thus automatically allowed to use deadly force to avoid being raped.
There is no legal difference between "firing to wound" and "firing to kill".
Basically, in a self-defense situation, you are not allowed to use deadly force unless you have a reasonable belief (i.e. the jury feels that in your shoes they'd feel the same way) that you were about to be subjected to deadly force. In this case, you can use deadly force (such as a gun) to prevent deadly force from being used against you. You *cannot* use deadly force if the threat is not immediate -- if the person is running away from you to get away, you cannot shoot them in the back.
If nondeadly force is used against you, you may only use nondeadly force to prevent it. If someone says that they're going to give you a wedgie, you can shove them away and run or knock them to the ground to get away, but you cannot spray them with an automatic weapon.
There is one exception to weapons being considered deadly force -- *threatening* someone with a gun is not considered deadly force. You *may* threaten a burglar with a weapon if you catch him in your house. However, unless he reaches for a pocket (possibly a gun) or rushes you, you cannot actually legally fire the weapon.
(There is an exception to this in Texas, where you are allowed to use deadly force against burglars.)
This could differ in Canda, but Canadian law and US law are awfully similar.
Some of the people in the world don't use Windows, you know?
Symantec doesn't make a virus scanner for Linux. So am I just screwed, because no Linux folks can examine viruses? How about AmigaOS?
Regardless of the fact, I see no reason why legislation should force me to buy one of Symantec's products anyway.
Banning the writing/possession/nonmalicious distribution of viruses would be just plain stupid.
The fact that I lived to see the day when someone can say this somewhat seriously is very upsetting to me.
Code is being put on a level with guns now.
:-(
I see your point, but lots of things directly do damage. The article author's point seemed to be that damage is the reason that virus writing/distribution is not okay.
So it comes down to whether we should ban something for practical reasons.
Guns and cars actually *kill* people, but they've been okayed under the don't-illegalize-it-unless-absolutely-necessary doctrine.
Frankly, the biggest malware threat I see is through worms spread by the technically illiterate -- in ICQ and email. I don't see redistribution of viruses as being all that big of a threat.
And I really dislike legislation against writing software, period. Make illegal the use of that software, okay, but not the writing of it. Code is just a little too close a the description of a thing. It feels like blueprints for bombs are being banned.
How about just *NOT USING THE NYT*?
Simple, easy, makes everyone happy. Link to other news sites in posts.
Ogle, a nice Linux DVD player, simply bypasses DVD protection attempts, according to my roommate. Not sure the technical details.
No one likes their damn accounts, no one likes being tracked, and there are a hell of a lot of other news sites. Please *stop* posting Slashdot articles with links to NYTimes articles. You just encourage them.
CMU's been doing a lot of work in developing nursing home care robots.
The most noticable thing about 'em is that they have these oversized facial features that grin or frown to try to get emotion across easily.
IRC is about the easiest to track form of communication I can think of, actually.
So hackers hang out on IRC. No shit. Most techies hang out on IRC at some point or another. I suppose CNN reporters don't generally do so...
The BBC's website kicks the shit out of CNN's. They've got a nice low graphics version.
I stopped using CNN for my main news source in favor of BBC (well, and Google Headlines) some time ago. CNN is really awful.
I thought it was kind of funny at the time, how many people who live in the US are using the BBC because most of the major US news sites are bad.
technews.com is where I go for tech business news, though.
Uh, huh. Doesn't work, eh?
Seems that you stopped telemarketing as soon as you could.
The more miserable telemarketers there are the harder it is to get people to stay to do the job.
Besides, making telemarketers as unhappy as they made you is downright cathartic.
The "I'm just doing it to earn money" thing doesn't fly -- or else drugrunning, assassination, etc. would be kosher as well. Find a way to earn money that doesn't make people unhappy and they won't be mean back.
With a private institution, you *can* do exactly this.
The KKK is rather unlikely to accept a Black Panther as a member.
Frankly, I don't really have any problem at all with this. I don't even have a problem with discriminating against chunks of the population at public workplaces -- sorry, but if you're a guy, you just can't work at Hooters. That's just how it is. If Hooters is making a really stupid decision, then they'll go out of business eventually as their competitors kick their butt.