Actually, the ALU's in Prescott don't even use a clock! It uses self resetting domino logic, so the speed is completely based on the manufacturing process and the speed of the transistors. Damn hard things to make, even harder to formally verify that they will always work, and as far as I know Intel is the only CPU manufacturer in the world to use something like this in a mass-produced product. So you can really say that the ALU is working at >12GHz or whatever since it isn't clocked. Oh, and Intel has been measuring their operation time in picoseconds for a while now;)
True, but I think the system the grandparent was promoting was using public funds to create a public solution, which still requires buying/paying for tools from the private sector. Instead of buying a "black box" and just trusting the company that made it to Do The Right Thing(tm), you buy the hardware from one company/group, pay another group to write the software with public funds (thus making the results open to the public so anyone can find problems/backdoors), and another group to actually run things. This is a great example of checks and balances: spreading power between many groups instead of just a few or only one, thus reducing the change of tyranny and power grabs. It's what a lot of our Constituion is based on, and I would welcome seeing the same happen to our voting system, seeing as how voting is the greatest power in the country.
Who says they have to use wireless??? I'm saying that wireless will be present in more and more devices around town and in those carried by their patients/visitors. Keeping 20-30 year old equipment around in face of spreading WiFi signals that causes them to stop working Just Because(tm) isn't a sound reason. WiFi will soon include MAN networks (networks that cover entire cities), and then they will have no choice but to replace devices that the 2.4GHz (or whatever frequency we will be using then) signals interfear with. Best practice would be to shield all the new stuff so the Next Great Thing doesn't require yet another upgrade do to interference.
It's my opinion that the equipment in places like hospitals should design themselves to be hit by a nuke and still function. If something like a little WiFi+VoIP causes an piece of equipment to stop working then the equipment needs to be replaces/reengineered, because WiFi is here to stay and VoIP is gaining momentum, so it will have to be done sooner or later.
I think/. is very quick to post bad news about SP2 because MS is out singing to the heavens about how much more "secure" it is and how they are taking security "seriously" now. Bugs like this are just evidence that MS is yet again trying to tack on security after-the-fact instead of doing what is necessary: start over and have security in mind from the ground up in designing, developing, and testing of their OS and applications. MS is still a breeding ground for viruses, backdoors, worms, etc, all because MS will not admit that their products are pieces of crap who's only positive traits are being easy to use (if you are already used to using MS products that is) and pretty to look at (if all you have ever seen is MS products). From a admistrative point of view, their stuff is buggy, bloated, and a POS at the source code level. The firewall in SP2 should be simple, clean, and not affected by ANY other program or hook instead of Windows. But MS couldn't even get that right, instead doing their usual "tie-it-into-eveything-else-we-loose-market-share- to-a-competitor" routine, giving the user a false sense of security. That is why/. railes against MS and SP2. We don't like being lied to by the marketing department.
Ya, let's provoke a nation that is under the leadership of a crazy guy who probably wouldn't bat an eye at killing all this citizens by launching a nuke at a bunch of US carriers or at Seoul. When the crazy guy has a gun you don't point your gun at him and start yelling at him. You talk very nicely and don't make any sudden moves until you are sure you can get the gun away from him before he shots you, himself, or any of the innocent people who happen to be nearby. We can't use the same tactics against NK that we did with USSR; at least they weren't crazy. Power hungry and a little mean, but not crazy like the leader of NK.
But how can you "steal" an idea? Do you no longer have that idea? Is that idea somehow less accessible to you now that someone else is "using" it? I think this mentality is what the article was trying to address: you can't use the same terms/concepts with IP as regular, physical property.
If I take your car from you, I've stolen it. I have deprived you of access and use of a physical object. Plus, there is the loss of the financial investment you made in the car. This is the regular idea of stealing.
Now say you come up with a need idea, like how to flap your arms just right so you can fly like a bird. If I "take" this idea and start flying around myself, have I deprived you of the knowledge on how to fly? Are you no longer able to benifit from this idea? Maybe you wanted to "sell" flying lessons at $499 a person. Now I teach people to fly at $299 a person instead. You can argue I've "stolen" money from you, but in reality I haven't taken any money from you that wasn already in your possession. If the people I teach to fly turn around and start teaching other people to fly for free, have they taken any money from you or I that we already had? In the end, you didn't make much money, I didn't make much money, but now we can all fly.
I think that people are so IP nuts is that they can't stand the idea of someone else benifiting from all the hardwork another person did without them having to put in any themselves. They feel cheated because they did the hard work and got the same rewards as the person who did none. But there are other benifits to work than just the almighty dollar. Recognition, respect, etc can be a great inspiration for doing this hard work. We all know who Linus is, and he didn't make any money (directly) from all his hard work and yet was Wired's "Leader of the Free World". He's left a mark on history that most of us will never equal. Sometimes that is worth more to a person than all the money in the world.
True, but the fact that XP (and 2K) relied on floppies for hd controller drivers post 1998 is kind of disconcerting. If they had programed it correctly it wouldn't have mattered what media (floppy, CD, DVD, etc) that the drivers came in.
1. I can install 3rd party disk drivers during a Windows install from a CD or USB device (right now you can only do that with a floppy)
2. EVERY BIOS supports booting from a USB key device
3. USB keys universally work across all platforms and OS's. Some do already, but some don't and rely on the OS to have builtin drivers already.
4. ALL OEMs stop relying on floppies for ANYTHING (Dell for example).
Once all these come to pass, we can safely throw away our floppies and be fine. Until then, floppies will cling to life by a thin thread for admins, hackers, and power users, even though none of them wish to use floppies. Normal users have no need for floppies these days, so this won't affect them much.
Well, I take it you haven't had much experience with OS programming then.
One key concept is the seperation of kernel space and user space. The kernel sits between the hardware and the programs. When a program wants to write something to the screen, it doesn't do it directly but throught the kernel. The kernel acts like a centry, so to speak, and prevents misuse or stealing of system resources by programs, and also provides a nice abstraction layer for the programers. The user's programs run in their own memory spaces (allocated by the kernel and enforced by the kernel) and the kernel sits in it's own protected memory space.
One of the biggest differences between MS and *nix is that the distinction between kernel space and user space is more blured in MS than in *nix. MS has had a bad habit of tying thier programs into deep hooks inside their kernel in order to get some extra functionality or a bit of performance increase. The problem with this is that it provides a route for errors/viruses/malware/etc to enter into the kernel space through a regular program. Hence, an IE bug becomes a Windows bug. If MS kept a better speration of their programs and thier kernel, a lot of thier problems would probably disappear.
I agree with you about security being a process, not something you can buy in a box. But what you get in that box determines where on the security spectrum you start. A system that allows any user to run as administrator and has a LOT of ports open by default (why I'll never know) is a bit harder to secure and keep secure than a system where there is a clear division between users, access rights, and power where things are turned off by default instead of on by default.
Doubt - On the touchy issue of security, Ballmer also dismissed the notion that Linux is more secure than Windows, saying that Linux would be attacked just as frequently as Windows if the open source operating system had as large a share of the operating system market as Windows.
I always love reading that one. Sure, it'll be attacked just as much, but like shooting a BB gun at a person in a suit of armor, they won't get through nearly as much as when shoot at that poor naked man (MS Windows).
Damn, I wish I had a mod point left so I could mod you +1 insightful or +1 informative.
This is one of the BIGGEST differences between Bush and Kerry. Bush is secretive, paranoid, and doesn't want to see/hear anything that he doesn't like/agree with. Kerry is more open and willing to suffer the chance of hecklers if it means more regular people can have contact with him.
I remember when they both came to my state (Oregon) a few weeks ago. Bush was in some school in Beaverton, which they didn't anounce until like the day before. Only people who where invited could meet with the President. Kerry had a big rally out on the waterfront (downtown Portland, very pretty and lots of room). Everyone could come and meet him. Personally, I want a President that will accessible to the people and not so paranoid about something bad happening that he has no real contact with the people he is supposed to represent.
Actually, it's the Dells of the world that prevent the Pentium M from being on the desktop. There is a premium for bying a Pentium M processor (not as massivily produced as P4's are), and the desktop manufacturers dont' want to pay that premium for a desktop chip. As for Intel releasing their own chipset/mobo for a desktop Pentium M, I think Intel believes there wouldn't be much of a market for them, so they don't produce them. But in no way is Intel "embarrassed" that the Pentium M is a good chip and thus are actively trying to keep it off the desktop. Money is money, and Intel (like any other business) will make money where they can.
If you check out most of Intel's latest mobile push (Centrino) you will notice that they are centered more on efficiency and power saving. I can run my Dothan based Centrino laptop for almost 4 hours and still get good performance out of it, even though the CPU is 1GHz slower than when at full speed. Intel can see that people want this, not faster and faster GHz (for a mobile anyway).
Doubtful. Each retical used in a fab is over $10,000 to make, and there are a LOT of reticals used to make the full chip. So, if you wanted a larger CPU, it wouldn't be 50% more, more like 500%. All the master layouts would have to be redone (which a lot are done by hand, which means $$$$$$$).
The problem is production, not cooling. By making the CPU die bigger, you a) decrease the number of dies you can make on a single wafer, which costs a fixed amount to produce, thus making each CPU more expensive; and b) defects that would have only scrapped 1 die out of 300 will now scrap 1 die out of 50, thus making the yields lower, raising the cost per die, making the CPU more expensive to the consumers. Decreasing the die size and increasing the wafer size leads to cheaper chips which is a Good Thing (tm). A nice side affect is that it also allows for higher clocking, which is both good (more ops per second) and bad (current leakage and heat issues). Smaller dies also consume less voltage, which is again a Good Thing (tm). Just have to get current leakage, a Bad Thing (tm), down and the chips would run cooler and consume less power. This new process is better at current leakage, so thats a Good Thing (tm). All in all, making the CPUs smaller is good for Intel and good for the consumer.
Um, why would they have to talk through the northbridge if they are physically in the same silicon? That's the whole point behind dual core chips: eliminate the communications issues found it current MP systems. If they are physically next to each other, why would they even think of going off chip to talk to its neighbor????????
Who says it's going to be $20? I'd buy it if it was $1.25 compared to teh $1. First for the coolness (no pun) factor, and second, I wouldn't have to worry about drinking it as soon as I got it. Buy it now, through it in my bag, drink it later when I'm thirty and have it self cool itself. Oh, and I don't know if any of you have had this experience, but putting an already chilled can in a bag causes massive amount of condensation to be left in the bag, getting everything wet. Being able to put in a room temperature can and not have to worry about everything getting wet is a plus in my book. But of course, this is all dependant on the cans only being slightly more expensive, not significantly.
I'll second that! How many more comparitive literature people do we really need these days? It's not like they have many job prospects or much to really contribute to society at large.
And this is why I don't bother with WEP. If using WEP is going to slow down my network for no real protection, I'll just go at full speed and not worry about securing it. Of course, I do MAC filter at my WAP, which does keep my network semi-closed.
Sunbird is a CALENDER system, not a browser. But you can download it as a plugin to the browser, which works really well:) It puts a Calender option under Tools. Very slick if you ask me.
Slashdot Mirror
Actually, the ALU's in Prescott don't even use a clock! It uses self resetting domino logic, so the speed is completely based on the manufacturing process and the speed of the transistors. Damn hard things to make, even harder to formally verify that they will always work, and as far as I know Intel is the only CPU manufacturer in the world to use something like this in a mass-produced product. So you can really say that the ALU is working at >12GHz or whatever since it isn't clocked. Oh, and Intel has been measuring their operation time in picoseconds for a while now ;)
True, but I think the system the grandparent was promoting was using public funds to create a public solution, which still requires buying/paying for tools from the private sector. Instead of buying a "black box" and just trusting the company that made it to Do The Right Thing(tm), you buy the hardware from one company/group, pay another group to write the software with public funds (thus making the results open to the public so anyone can find problems/backdoors), and another group to actually run things. This is a great example of checks and balances: spreading power between many groups instead of just a few or only one, thus reducing the change of tyranny and power grabs. It's what a lot of our Constituion is based on, and I would welcome seeing the same happen to our voting system, seeing as how voting is the greatest power in the country.
Who says they have to use wireless??? I'm saying that wireless will be present in more and more devices around town and in those carried by their patients/visitors. Keeping 20-30 year old equipment around in face of spreading WiFi signals that causes them to stop working Just Because(tm) isn't a sound reason. WiFi will soon include MAN networks (networks that cover entire cities), and then they will have no choice but to replace devices that the 2.4GHz (or whatever frequency we will be using then) signals interfear with. Best practice would be to shield all the new stuff so the Next Great Thing doesn't require yet another upgrade do to interference.
It's my opinion that the equipment in places like hospitals should design themselves to be hit by a nuke and still function. If something like a little WiFi+VoIP causes an piece of equipment to stop working then the equipment needs to be replaces/reengineered, because WiFi is here to stay and VoIP is gaining momentum, so it will have to be done sooner or later.
I think /. is very quick to post bad news about SP2 because MS is out singing to the heavens about how much more "secure" it is and how they are taking security "seriously" now. Bugs like this are just evidence that MS is yet again trying to tack on security after-the-fact instead of doing what is necessary: start over and have security in mind from the ground up in designing, developing, and testing of their OS and applications. MS is still a breeding ground for viruses, backdoors, worms, etc, all because MS will not admit that their products are pieces of crap who's only positive traits are being easy to use (if you are already used to using MS products that is) and pretty to look at (if all you have ever seen is MS products). From a admistrative point of view, their stuff is buggy, bloated, and a POS at the source code level. The firewall in SP2 should be simple, clean, and not affected by ANY other program or hook instead of Windows. But MS couldn't even get that right, instead doing their usual "tie-it-into-eveything-else-we-loose-market-share- to-a-competitor" routine, giving the user a false sense of security. That is why /. railes against MS and SP2. We don't like being lied to by the marketing department.
Ya, let's provoke a nation that is under the leadership of a crazy guy who probably wouldn't bat an eye at killing all this citizens by launching a nuke at a bunch of US carriers or at Seoul. When the crazy guy has a gun you don't point your gun at him and start yelling at him. You talk very nicely and don't make any sudden moves until you are sure you can get the gun away from him before he shots you, himself, or any of the innocent people who happen to be nearby. We can't use the same tactics against NK that we did with USSR; at least they weren't crazy. Power hungry and a little mean, but not crazy like the leader of NK.
If I take your car from you, I've stolen it. I have deprived you of access and use of a physical object. Plus, there is the loss of the financial investment you made in the car. This is the regular idea of stealing.
Now say you come up with a need idea, like how to flap your arms just right so you can fly like a bird. If I "take" this idea and start flying around myself, have I deprived you of the knowledge on how to fly? Are you no longer able to benifit from this idea? Maybe you wanted to "sell" flying lessons at $499 a person. Now I teach people to fly at $299 a person instead. You can argue I've "stolen" money from you, but in reality I haven't taken any money from you that wasn already in your possession. If the people I teach to fly turn around and start teaching other people to fly for free, have they taken any money from you or I that we already had? In the end, you didn't make much money, I didn't make much money, but now we can all fly.
I think that people are so IP nuts is that they can't stand the idea of someone else benifiting from all the hardwork another person did without them having to put in any themselves. They feel cheated because they did the hard work and got the same rewards as the person who did none. But there are other benifits to work than just the almighty dollar. Recognition, respect, etc can be a great inspiration for doing this hard work. We all know who Linus is, and he didn't make any money (directly) from all his hard work and yet was Wired's "Leader of the Free World". He's left a mark on history that most of us will never equal. Sometimes that is worth more to a person than all the money in the world.
Are they going to call it Kermit the Bot?
True, but the fact that XP (and 2K) relied on floppies for hd controller drivers post 1998 is kind of disconcerting. If they had programed it correctly it wouldn't have mattered what media (floppy, CD, DVD, etc) that the drivers came in.
1. I can install 3rd party disk drivers during a Windows install from a CD or USB device (right now you can only do that with a floppy) 2. EVERY BIOS supports booting from a USB key device 3. USB keys universally work across all platforms and OS's. Some do already, but some don't and rely on the OS to have builtin drivers already. 4. ALL OEMs stop relying on floppies for ANYTHING (Dell for example). Once all these come to pass, we can safely throw away our floppies and be fine. Until then, floppies will cling to life by a thin thread for admins, hackers, and power users, even though none of them wish to use floppies. Normal users have no need for floppies these days, so this won't affect them much.
One key concept is the seperation of kernel space and user space. The kernel sits between the hardware and the programs. When a program wants to write something to the screen, it doesn't do it directly but throught the kernel. The kernel acts like a centry, so to speak, and prevents misuse or stealing of system resources by programs, and also provides a nice abstraction layer for the programers. The user's programs run in their own memory spaces (allocated by the kernel and enforced by the kernel) and the kernel sits in it's own protected memory space.
One of the biggest differences between MS and *nix is that the distinction between kernel space and user space is more blured in MS than in *nix. MS has had a bad habit of tying thier programs into deep hooks inside their kernel in order to get some extra functionality or a bit of performance increase. The problem with this is that it provides a route for errors/viruses/malware/etc to enter into the kernel space through a regular program. Hence, an IE bug becomes a Windows bug. If MS kept a better speration of their programs and thier kernel, a lot of thier problems would probably disappear.
I agree with you about security being a process, not something you can buy in a box. But what you get in that box determines where on the security spectrum you start. A system that allows any user to run as administrator and has a LOT of ports open by default (why I'll never know) is a bit harder to secure and keep secure than a system where there is a clear division between users, access rights, and power where things are turned off by default instead of on by default.
I always love reading that one. Sure, it'll be attacked just as much, but like shooting a BB gun at a person in a suit of armor, they won't get through nearly as much as when shoot at that poor naked man (MS Windows).
Damn, I wish I had a mod point left so I could mod you +1 insightful or +1 informative.
This is one of the BIGGEST differences between Bush and Kerry. Bush is secretive, paranoid, and doesn't want to see/hear anything that he doesn't like/agree with. Kerry is more open and willing to suffer the chance of hecklers if it means more regular people can have contact with him.
I remember when they both came to my state (Oregon) a few weeks ago. Bush was in some school in Beaverton, which they didn't anounce until like the day before. Only people who where invited could meet with the President. Kerry had a big rally out on the waterfront (downtown Portland, very pretty and lots of room). Everyone could come and meet him. Personally, I want a President that will accessible to the people and not so paranoid about something bad happening that he has no real contact with the people he is supposed to represent.
Actually, it's the Dells of the world that prevent the Pentium M from being on the desktop. There is a premium for bying a Pentium M processor (not as massivily produced as P4's are), and the desktop manufacturers dont' want to pay that premium for a desktop chip. As for Intel releasing their own chipset/mobo for a desktop Pentium M, I think Intel believes there wouldn't be much of a market for them, so they don't produce them. But in no way is Intel "embarrassed" that the Pentium M is a good chip and thus are actively trying to keep it off the desktop. Money is money, and Intel (like any other business) will make money where they can.
If you check out most of Intel's latest mobile push (Centrino) you will notice that they are centered more on efficiency and power saving. I can run my Dothan based Centrino laptop for almost 4 hours and still get good performance out of it, even though the CPU is 1GHz slower than when at full speed. Intel can see that people want this, not faster and faster GHz (for a mobile anyway).
Doubtful. Each retical used in a fab is over $10,000 to make, and there are a LOT of reticals used to make the full chip. So, if you wanted a larger CPU, it wouldn't be 50% more, more like 500%. All the master layouts would have to be redone (which a lot are done by hand, which means $$$$$$$).
True, but when you are only at 20%, you still have a LOOOOOOONG way to go. You can start complaining about this when MS is closer to 95-99% :-P
But that's for talking to OFF CHIP memory, not on chip memory such as cache. Guess you don't understand that concept.
The problem is production, not cooling. By making the CPU die bigger, you a) decrease the number of dies you can make on a single wafer, which costs a fixed amount to produce, thus making each CPU more expensive; and b) defects that would have only scrapped 1 die out of 300 will now scrap 1 die out of 50, thus making the yields lower, raising the cost per die, making the CPU more expensive to the consumers. Decreasing the die size and increasing the wafer size leads to cheaper chips which is a Good Thing (tm). A nice side affect is that it also allows for higher clocking, which is both good (more ops per second) and bad (current leakage and heat issues). Smaller dies also consume less voltage, which is again a Good Thing (tm). Just have to get current leakage, a Bad Thing (tm), down and the chips would run cooler and consume less power. This new process is better at current leakage, so thats a Good Thing (tm). All in all, making the CPUs smaller is good for Intel and good for the consumer.
Um, why would they have to talk through the northbridge if they are physically in the same silicon? That's the whole point behind dual core chips: eliminate the communications issues found it current MP systems. If they are physically next to each other, why would they even think of going off chip to talk to its neighbor????????
You know, you might want to hide your porn better in the future then ;)
Who says it's going to be $20? I'd buy it if it was $1.25 compared to teh $1. First for the coolness (no pun) factor, and second, I wouldn't have to worry about drinking it as soon as I got it. Buy it now, through it in my bag, drink it later when I'm thirty and have it self cool itself. Oh, and I don't know if any of you have had this experience, but putting an already chilled can in a bag causes massive amount of condensation to be left in the bag, getting everything wet. Being able to put in a room temperature can and not have to worry about everything getting wet is a plus in my book. But of course, this is all dependant on the cans only being slightly more expensive, not significantly.
I'll second that! How many more comparitive literature people do we really need these days? It's not like they have many job prospects or much to really contribute to society at large.
And this is why I don't bother with WEP. If using WEP is going to slow down my network for no real protection, I'll just go at full speed and not worry about securing it. Of course, I do MAC filter at my WAP, which does keep my network semi-closed.
Sunbird is a CALENDER system, not a browser. But you can download it as a plugin to the browser, which works really well :) It puts a Calender option under Tools. Very slick if you ask me.