WTF? Why would Intel care what OS you run? Hell for quite a while after Itanium shipped the only non-beta quality OS to run on it was Linux. Now to placate the content industry Intel is working on DRM to protect content, but that only means that you might not be able to play certain content under Other_OS, or that it might require Non_Free and Binary_Only App to view. Being a little paranoid is often healthy, but being very paranoid and uninformed is not.
The real problem is on the consumption side, not the supply side. The issue is that (at least in the US, and probably also in Europe) we consume more energy than is currently falling on the land mass we occupy. If you look into farming oil producing plants without fuel derived fertilizers, you quickly realize that there simply isn't enough solar energy hitting the ground to power our current usage patterns even without real world inefficiencies. The total solar energy falling on the earths arable landmass (18 million square km, or about 12% of the total surface) is around 20,000 TW/hour (average of 200W/m^2/hour), so assuming 100% efficiency and exclusive use of arable land for energy production you get a total energy input of only 25kWh per person while the sun shines based on 8 billion humans. The average old american house has 100A 120V service, which means they are capable of pulling 12kWh, with newer homes going as high as 400A or 48kWh. This means that many large american homes would utilize more than their share of the worlds solar energy input just for the home with 100% energy transfer efficiency. Counting in the family cars and the energy needed to produce and ship the goods they consume including food and you see that we in the west are really consuming way more energy than is sustainable!
Your "observations" totally overlook one of the areas where hybrids gain a lot of their efficiency, by running the engine in the optimal portion of the power band. Also in most of today's hybrid systems they get away with such aenemic engines by supplementing the horsepower with the electric motors, if you add 100+HP from the electric system when accelerating then who cares if your engine is underpowered, other than accelerating the only thing that HP helps with is top speed, which is governed by law for 99.99+% of the population anyways.
Ground level Ozone is BAD, it makes it much harder to breathe and can cause asthmatic's to have severe attacks, often leading to death. The ozone produced as part of fuel consumption rarely makes it into the upper atmosphere where it would help to filter UV rays. Almost all of the air quality alerts where I live are due to elevated ozone levels, not high particulate counts.
WEP and pre 802.1x wireless security regimens ARE worthless. 128 bit WEP can be broken in less than 4 minutes as linked to from slashdot. The only way I can see allowing wireless on my network is if I get to treat it just like the internet, which means only limited access to specific hosts on specific ports. And of course to enforce that kind of security I can't have random people plugging in their SOHO devices, so that leads back to the articles main question, how do you identify non-approved AP's =)
If you were truely BOFHly you would just turn the port down administrativly, no sense in walking into the boobytrapped datacenter if you don't have to =)
Not sure where you live but many roadside billboards around Cleveland are already going to animated and/or changing rotations. Aparantly someone has come up with a Jumbotron like technology with fairly low glare and auto-adjustment for lighting conditions which is cheap enough to use for billboards. I personally find it really annoying, but now that I know where they are I just ignore them when I see the flash of animation out of the corner of my eye.
HP, Compaq, Dell, Gateway, Toshiba. Those five account for over 2/3rds of the PC market. None of them was going to hand their rival a large revenue stream if it could be avoided.
Yep, got to love NonStop/Himalaya, it's the only non mainframe platform that I've used that comes close to the uptime and data integrity requirements of the mainframe world.
I wonder if IBM supports the crypto card that is at the heart of the ATM network under Linux? Because without support for those cards there is zero chance of Linux replacing OS/2 in the ATM space.
OS/2 never had a chance, specifically because it came from IBM. No way in hell were the other major OEM's going to feed their biggest competitor by buying the OS from IBM.
About a year and a half ago I was part of a VERY large re-rollout of OS/2 at Washington Mutual. They had been running two boxes with KVM's for a large %age of their loan processors because they had some custom apps that were designed for OS/2. We took out the second box and upped the ram in the windows boxes. Then we installed a new XP image that included Virtual PC with an OS/2 image. This install was nationwide and included thousands and thousands of seats (hell just in Northridge they have a campus that's FOUR city blocks!)
Yep, it's broken, but Rob doesn't care. Since the system reduces the amount of bot crap coming from compromised hosts he thinks it's acceptable to portscan you every time you post a comment. Personally I find it annoying and offensive, but not enough so to stop posting to slashdot.
The L in LDAP comes from the fact that it was a stripped down version of the full X.500 directory schema. While LDAP is a big hairy beast of a standard it is NOWHERE near as complex to code or admin as a full X.500 implementation would be.
Is the SFU server really just NIS, or is it a NIS+ server. Because if it's simply NIS, that is BAD, craptacular sercurity and no mutual authentication make this sysadmin a sad boy.
Ok, so you have a single core PPC 970FX at 16W, or a dual core Pentium M Yonah at 25W. I can't find any numbers for the PPC 970MP, but I'm assuming it will be in the neighborhood of 1.5-2x the draw of the PPC 970FX low power since that chip already uses all of IBM's current tricks. Add to that the fact that IBM kept dragging their feet on any PPC 970 development and focused all their energy on the console market while Apple was left for 18 months with a stagnant product line and I can understand why they made the switch.
Uh, the only way those charts would not show Intel kicking arse and taking names is if you intentionally left the Pentium M architecture off. With dual core Pentium M's at fast speeds on the horizon everything but they very top of the market can be serviced by a CPU with a typical power consumption of 1 Watt and a max consumption of only 25W.
Of COURSE it's the webmasters responsibility to opt out! If you put something up on a publically available website then you can expect the public to consume and possibly retain copies of your work. Heck failing to opt out and then suing is like suing a library because they have an old copy of your book on hand. It's no big secret that there are websites out there which use spiders to download, store, and analyze the vast majority of the internet, and that your site will be included in such efforts unless you purposly decide to disallow such use through the commonly agreed upon methods.
The idea, at least as far as what I have generally pushed is that you improve security (no data except on the servers), you reduce the need for constant upgrade cycles (go from replacing desktops every 3 years to replacing thins every ~10), and finally the biggy, managability (someones thin dies, all you do is replace it with another thin, they get all their same apps, all their customization is still in place, etc without having to do ANY additional work). Plus it's easier to maintain a small farm or two of servers then it is to manage a fleet of desktops, no matter how good the management tools are for the desktops.
I assume that they hooked the Windows CPUID function and returned Genuine Intel blah instead of the real CPUID return from the CPU, that or they patched the object file to always use the Intel codepath.
Some of the HP Professional series inkjet printers already can do real world 20+ ppm for black text, with 4x the nozels I believe they could hit somewhere near their claimed speed for simple documents. Heck the Laserjet line now goes up to 45ppm for the black and white 4250 line, and yes I really get those kind of print speeds.
Because the resolution and color saturation on even the best color laser blows compared to a $100 inkjet from 5 years ago? Now if you want to throw thermal wax transfer and dye sublimation into the mix then yes there are better options, but they aren't as cheap as inkjet (either per copy or upfront).
Slashdot Mirror
WTF? Why would Intel care what OS you run? Hell for quite a while after Itanium shipped the only non-beta quality OS to run on it was Linux. Now to placate the content industry Intel is working on DRM to protect content, but that only means that you might not be able to play certain content under Other_OS, or that it might require Non_Free and Binary_Only App to view. Being a little paranoid is often healthy, but being very paranoid and uninformed is not.
The real problem is on the consumption side, not the supply side. The issue is that (at least in the US, and probably also in Europe) we consume more energy than is currently falling on the land mass we occupy. If you look into farming oil producing plants without fuel derived fertilizers, you quickly realize that there simply isn't enough solar energy hitting the ground to power our current usage patterns even without real world inefficiencies. The total solar energy falling on the earths arable landmass (18 million square km, or about 12% of the total surface) is around 20,000 TW/hour (average of 200W/m^2/hour), so assuming 100% efficiency and exclusive use of arable land for energy production you get a total energy input of only 25kWh per person while the sun shines based on 8 billion humans. The average old american house has 100A 120V service, which means they are capable of pulling 12kWh, with newer homes going as high as 400A or 48kWh. This means that many large american homes would utilize more than their share of the worlds solar energy input just for the home with 100% energy transfer efficiency. Counting in the family cars and the energy needed to produce and ship the goods they consume including food and you see that we in the west are really consuming way more energy than is sustainable!
Your "observations" totally overlook one of the areas where hybrids gain a lot of their efficiency, by running the engine in the optimal portion of the power band. Also in most of today's hybrid systems they get away with such aenemic engines by supplementing the horsepower with the electric motors, if you add 100+HP from the electric system when accelerating then who cares if your engine is underpowered, other than accelerating the only thing that HP helps with is top speed, which is governed by law for 99.99+% of the population anyways.
Ground level Ozone is BAD, it makes it much harder to breathe and can cause asthmatic's to have severe attacks, often leading to death. The ozone produced as part of fuel consumption rarely makes it into the upper atmosphere where it would help to filter UV rays. Almost all of the air quality alerts where I live are due to elevated ozone levels, not high particulate counts.
WEP and pre 802.1x wireless security regimens ARE worthless. 128 bit WEP can be broken in less than 4 minutes as linked to from slashdot. The only way I can see allowing wireless on my network is if I get to treat it just like the internet, which means only limited access to specific hosts on specific ports. And of course to enforce that kind of security I can't have random people plugging in their SOHO devices, so that leads back to the articles main question, how do you identify non-approved AP's =)
If you were truely BOFHly you would just turn the port down administrativly, no sense in walking into the boobytrapped datacenter if you don't have to =)
What, windowskey-d is too hard for you?
Not sure where you live but many roadside billboards around Cleveland are already going to animated and/or changing rotations. Aparantly someone has come up with a Jumbotron like technology with fairly low glare and auto-adjustment for lighting conditions which is cheap enough to use for billboards. I personally find it really annoying, but now that I know where they are I just ignore them when I see the flash of animation out of the corner of my eye.
HP, Compaq, Dell, Gateway, Toshiba. Those five account for over 2/3rds of the PC market. None of them was going to hand their rival a large revenue stream if it could be avoided.
Yep, got to love NonStop/Himalaya, it's the only non mainframe platform that I've used that comes close to the uptime and data integrity requirements of the mainframe world.
I wonder if IBM supports the crypto card that is at the heart of the ATM network under Linux? Because without support for those cards there is zero chance of Linux replacing OS/2 in the ATM space.
OS/2 never had a chance, specifically because it came from IBM. No way in hell were the other major OEM's going to feed their biggest competitor by buying the OS from IBM.
About a year and a half ago I was part of a VERY large re-rollout of OS/2 at Washington Mutual. They had been running two boxes with KVM's for a large %age of their loan processors because they had some custom apps that were designed for OS/2. We took out the second box and upped the ram in the windows boxes. Then we installed a new XP image that included Virtual PC with an OS/2 image. This install was nationwide and included thousands and thousands of seats (hell just in Northridge they have a campus that's FOUR city blocks!)
Yep, it's broken, but Rob doesn't care. Since the system reduces the amount of bot crap coming from compromised hosts he thinks it's acceptable to portscan you every time you post a comment. Personally I find it annoying and offensive, but not enough so to stop posting to slashdot.
The L in LDAP comes from the fact that it was a stripped down version of the full X.500 directory schema. While LDAP is a big hairy beast of a standard it is NOWHERE near as complex to code or admin as a full X.500 implementation would be.
Yes, and as the Windows 2000 Server splashscreen keeps reminding me it's built on NT (new technology) technology!
Is the SFU server really just NIS, or is it a NIS+ server. Because if it's simply NIS, that is BAD, craptacular sercurity and no mutual authentication make this sysadmin a sad boy.
The newer Tektronix scopes can be connected to the LAN, but they are a bit more than $20K =)
Ok, so you have a single core PPC 970FX at 16W, or a dual core Pentium M Yonah at 25W. I can't find any numbers for the PPC 970MP, but I'm assuming it will be in the neighborhood of 1.5-2x the draw of the PPC 970FX low power since that chip already uses all of IBM's current tricks. Add to that the fact that IBM kept dragging their feet on any PPC 970 development and focused all their energy on the console market while Apple was left for 18 months with a stagnant product line and I can understand why they made the switch.
Uh, the only way those charts would not show Intel kicking arse and taking names is if you intentionally left the Pentium M architecture off. With dual core Pentium M's at fast speeds on the horizon everything but they very top of the market can be serviced by a CPU with a typical power consumption of 1 Watt and a max consumption of only 25W.
Of COURSE it's the webmasters responsibility to opt out! If you put something up on a publically available website then you can expect the public to consume and possibly retain copies of your work. Heck failing to opt out and then suing is like suing a library because they have an old copy of your book on hand. It's no big secret that there are websites out there which use spiders to download, store, and analyze the vast majority of the internet, and that your site will be included in such efforts unless you purposly decide to disallow such use through the commonly agreed upon methods.
The idea, at least as far as what I have generally pushed is that you improve security (no data except on the servers), you reduce the need for constant upgrade cycles (go from replacing desktops every 3 years to replacing thins every ~10), and finally the biggy, managability (someones thin dies, all you do is replace it with another thin, they get all their same apps, all their customization is still in place, etc without having to do ANY additional work). Plus it's easier to maintain a small farm or two of servers then it is to manage a fleet of desktops, no matter how good the management tools are for the desktops.
I assume that they hooked the Windows CPUID function and returned Genuine Intel blah instead of the real CPUID return from the CPU, that or they patched the object file to always use the Intel codepath.
Some of the HP Professional series inkjet printers already can do real world 20+ ppm for black text, with 4x the nozels I believe they could hit somewhere near their claimed speed for simple documents. Heck the Laserjet line now goes up to 45ppm for the black and white 4250 line, and yes I really get those kind of print speeds.
Because the resolution and color saturation on even the best color laser blows compared to a $100 inkjet from 5 years ago? Now if you want to throw thermal wax transfer and dye sublimation into the mix then yes there are better options, but they aren't as cheap as inkjet (either per copy or upfront).