The overall cooling is better: one server room with real cooling is much more efficient than scattering the cooling throughout the building, and more reliable when the management turns off the cooling for the cubicles over the weekend.
The individual blade and cooliing management can be made quite a lot better: blades are notoriously easier to seriously vent than are desktop chassis's with their tangles of randomly located components and cables.
You've neglected physical management and repair. The nightmare of the variety of desktop clients and where they're stuffed in people's desktops is a source of incredible pain for anyone in IT. The ability to keep a few spare blades and swap them in is a lifesaver to restoring people's systems.
UPS management, cooling, and auditing the machines for accounting or security reasons all benefit, and you're much less likely to have employees steal RAM or CD drives when you're not looking. The problem is that they're more expensive for several reasons, and people won't have access to the CD drives or USB ports to plug in their CD's and play music or games or even install software when needed.
If you're going to make this kind of claim, I'd like to see the numbers, particularly of false negatives. (Where the real user fails to be identified by a system set to be picky enough to reject casual fakery.)
I have difficulty believing your claim: I can believe the manufacturer makes the claim and does a demo, but I want to see it with the Gummy Fingers described elsewhere.
It's about the same as the state for speech recognition elsewhere. The systems use way too little data to actually analyze and get at best a 95% or so recognition of the acutal user, and the sensor acuity to defeat even the fake gelatin fingers (Google keyword: gummi fingers) is simply not there, since with a fake finger made from a fingerprint lifted from elsewhere the class that did the Gummi fingers still got better than 80% recognition.
Basically, the ability to detect a fake fingerprint with a casual test has never existed. The sensors just aren't good enough, even if the software authors were willing to invest the resources to store really thorough images of fingerprints, which they're not.
Go to the interview. Then walk out citing their criminal behavior as making them unsafe and unsuitable employers for an ethical person. This game is even more fun if they actually offer you the job, which you might use to leverage up your current salary elsewhere.
Make them waste the money and recruiter time, and make their managers be aware that their unethical and criminal behavior costs them. Of course, you might lose your friends' good will if you do this. But given their employer's theft and abuse, can you trust those friends, or are they fighting the good fight from inside enemy lines?
Remember, when you sign an NDA to discuss your work with Microsoft, they can and will refuse the deal, steal your work, and bankrupt you to block your competition by undercutting you. There's plenty of intellectual property left for them to steal or lock up with their "embrace and extend" approach.
Using the brains and money on the legal team to defend against the lawsuits can be more money-making than hiring developers: developers are very expensive.
Of course not. They're not offering to pay me to think like a PhD, and in fact the pay for being a PhD sucks. If I were applying for a degree, or as part of their research lab, I'd expect such questions.
In addition, there's a gambler's chance that SCO will win the case. It's enough for SCO to go to investors and lie, much like a lot of dotcom business plans. And they've clearly had a lot of practice lying to investors, because they're hiding as much as possible the amount of their business that remains because Microsoft is propping them up.
Microsoft is not taking over SCO or themselves directly providing the business because it would show Microsoft sponsorship of the lawsuit. But investing the money through 3-way partnerships and providing "partner" support is a very, very easy way, and much cheaper than doing lawsuits themselves, for Microsoft to cast doubts on the legitimacy of Linux.
By providing less than enough money for the lawsuits, Microsoft also helps keep SCO, the current owner of the UNIX trademark and source code, crippled as well and interferes with UNIX development. Who wants to use a core OS owned by such an idiot company, and with the SCO/Novell argument casting doubts on the legitimacy of its ownership? So developers remain fragmented with FreeBSD, NetBSD, Linux, etc. as only a few companies remain willing to buy or use the core UNIX source, which actually had good stuff in it.
You're comparing apples and oranges and leaving out the orange. "I had to install Quake3, and it said to go get the updated drivers, so I did, but then my screen would only run in 640x480, and the virus software started complaining, and MS Office wouldn't run!"
No kidding, this is what happened on my game system. It took several rounds of updates and yanking out my good video card and using a cheap one to get the machine bootable with the resulting mess.
OK, Star Trek transporters ws hyperbole, but not by much. Building a large-scale mass driver is still wildly infeasible: the switching of large magnetic fields at high propagation rates has turned on to be far more difficult than anticipated when mass drivers were first proposed. Like computers that understand spoken English, the proof of concept is easy. Getting it to work has turned out to be a bear.
That is not Rutan's fault. The difficulties he faced were not so much physical as political, with NASA's blatant and covert interference with his project.
Take a look at the numbers. Rutan and his cohorts could replace the entire Shuttle operation in 2 years with half the annual budget, if NASA would fire the bureaucrats who are actively interfering with his project. The basic technologies of his design are more modern, simpler to implement, and a prime example of a "skunkworks" project of brillienat people with a good budget and excellent managers not interested in building fiefdoms at the expense of the project.
The short answer is "no". No one has ever built an effective mass driver: the technology is about as available as Star Trek transporters, and should be considered as likely to work until someone actually does the design and testing of even a small one for launching significant masses across a few hundred feet, much less launching to LEO.
Agreed about the corporate welfare payments, but don't forget the bureaucracy welfare payments to NASA's current management staff.
Dump NASA, or get a decent mandate in place to get out of the way of the Phoenix and its developers, and watch a Shuttle built the right way within 10 years. I'm wildly impressed by the Phoenix developers, they're doing all their basic engineering and most of their bureaucracy correctly.
Unfortunately, a lot of old systems don't have bootable USB accessible without yanking the system away from the wall and plugging in the back, or use weird chipsets that it's hard to build kernels to deal with.
The live bootable CD's and DVD's work very well because booting from the built-in CD or DVD is something any modern system absolutely must support. They're also very useful in that you can't write anything on them, and this helps prevent all sorts of potential security concerns, unlike the bootable USB devices.
Look, the OS really doesn't matter. What does matter is getting your employers to not do stupid things, like run their laptops without security patches and insist on running NFS and file sharing from home and on every machine in your group, getting them to pick decent passwords, teaching people never to use.zip attachments for anything, never running passphraseless accounts and open access points, etc., etc., etc.
Until you can get basic security steps like those in place, the world's best firewall is like a really big lock on a 3 foot high fence. Even the most casual crackers will simply step over it.
Install Linux on hardware that is decommissioned for running Windows, and on which you don't have the licenses to run Windows. Especially do this on your desktop machine to get the debugging tools and viral immunity, and use the remote desktop tools to log into Windows Pro boxes as necessary to administer them remotely.
Then stop buying new Windows servers and make faster, more secure servers out of the decommissioned Windows boxes. And enjoy.
GOOD FOR YOU! Knowledgable, competent cross-platform work is a blessing and should be encouraged.
They see you as encroaching on their turf, rather than filling their needs by getting them the other resources they need. It drives the fiefdom leaders nuts when you point out how to save computers, money, and licenses by using superior tools that didn't come from their consultants or their "approved vendor", because it didn't come from their "process".
Establishing the trust needed to be given information, and to share ideas, in those other fields is vital and can be hard. So many outsiders to the field give such half-assed advice that people can set their filters very high to keep from having their time wasted. That's a human skill, and a precious one.
The future is the Trusted Computing Initiative: the activation codes will be node-locked to the BIOS or the CPU, with decently well encrypted keys, and you had better believe that these systems will be calling back to the mothership with your locally generated public keys to verify your ownership of your software, or even your ability to boot the machine because you've purchased an OS with a registered kernel.
Graceful key management without having to staple in tools like "keychains", a nice UI, really good VT100 and other terminal emulation that is vatly superior to that of the CygWin command prompt or the Windows telnet tools, good SSH tunneling management, and someone you can call at 3 AM with your software problem to walk you through debugging the issues. SecureCRT is a very sweet tool, and I recommend it.
Now, if you compared it to *Putty*, the open source OpenSSH based free tool built for Windows, you'd have a better basis for comparison.
If you're a piracy crook, you've got remote access to someone else's machine, possibly to hundreds of them. You build, or find, a pirate software website that publishes these.
The invested time is really not very large and scales nicely.
They also don't have the local bandwidth requirements of, say, a banking facility processing stock predictions and transactions that needs to pass many Gigabytes of data among a local cluster. Their bandwidth is more limited by their external access, and usage by the last mile of cable to their user's machines.
Good hotswap and RAID5 is expensive, especially if you want to buy the good 3Ware or Adaptec stuff instead of that abomiable and undocumented Promise stuff you see in desktop motherboards.
Slashdot Mirror
I bless Dell for this. They more than pay for themselves with saving IT time and downtime.
The overall cooling is better: one server room with real cooling is much more efficient than scattering the cooling throughout the building, and more reliable when the management turns off the cooling for the cubicles over the weekend. The individual blade and cooliing management can be made quite a lot better: blades are notoriously easier to seriously vent than are desktop chassis's with their tangles of randomly located components and cables.
You've neglected physical management and repair. The nightmare of the variety of desktop clients and where they're stuffed in people's desktops is a source of incredible pain for anyone in IT. The ability to keep a few spare blades and swap them in is a lifesaver to restoring people's systems. UPS management, cooling, and auditing the machines for accounting or security reasons all benefit, and you're much less likely to have employees steal RAM or CD drives when you're not looking. The problem is that they're more expensive for several reasons, and people won't have access to the CD drives or USB ports to plug in their CD's and play music or games or even install software when needed.
If you're going to make this kind of claim, I'd like to see the numbers, particularly of false negatives. (Where the real user fails to be identified by a system set to be picky enough to reject casual fakery.)
I have difficulty believing your claim: I can believe the manufacturer makes the claim and does a demo, but I want to see it with the Gummy Fingers described elsewhere.
It's about the same as the state for speech recognition elsewhere. The systems use way too little data to actually analyze and get at best a 95% or so recognition of the acutal user, and the sensor acuity to defeat even the fake gelatin fingers (Google keyword: gummi fingers) is simply not there, since with a fake finger made from a fingerprint lifted from elsewhere the class that did the Gummi fingers still got better than 80% recognition.
Basically, the ability to detect a fake fingerprint with a casual test has never existed. The sensors just aren't good enough, even if the software authors were willing to invest the resources to store really thorough images of fingerprints, which they're not.
Go to the interview. Then walk out citing their criminal behavior as making them unsafe and unsuitable employers for an ethical person. This game is even more fun if they actually offer you the job, which you might use to leverage up your current salary elsewhere. Make them waste the money and recruiter time, and make their managers be aware that their unethical and criminal behavior costs them. Of course, you might lose your friends' good will if you do this. But given their employer's theft and abuse, can you trust those friends, or are they fighting the good fight from inside enemy lines?
Remember, when you sign an NDA to discuss your work with Microsoft, they can and will refuse the deal, steal your work, and bankrupt you to block your competition by undercutting you. There's plenty of intellectual property left for them to steal or lock up with their "embrace and extend" approach. Using the brains and money on the legal team to defend against the lawsuits can be more money-making than hiring developers: developers are very expensive.
Of course not. They're not offering to pay me to think like a PhD, and in fact the pay for being a PhD sucks. If I were applying for a degree, or as part of their research lab, I'd expect such questions.
In addition, there's a gambler's chance that SCO will win the case. It's enough for SCO to go to investors and lie, much like a lot of dotcom business plans. And they've clearly had a lot of practice lying to investors, because they're hiding as much as possible the amount of their business that remains because Microsoft is propping them up.
Microsoft is not taking over SCO or themselves directly providing the business because it would show Microsoft sponsorship of the lawsuit. But investing the money through 3-way partnerships and providing "partner" support is a very, very easy way, and much cheaper than doing lawsuits themselves, for Microsoft to cast doubts on the legitimacy of Linux.
By providing less than enough money for the lawsuits, Microsoft also helps keep SCO, the current owner of the UNIX trademark and source code, crippled as well and interferes with UNIX development. Who wants to use a core OS owned by such an idiot company, and with the SCO/Novell argument casting doubts on the legitimacy of its ownership? So developers remain fragmented with FreeBSD, NetBSD, Linux, etc. as only a few companies remain willing to buy or use the core UNIX source, which actually had good stuff in it.
Oh, please. The encap package has been beating out stow for actual usability for years. Just read the docs on both for a comparison.
You're comparing apples and oranges and leaving out the orange. "I had to install Quake3, and it said to go get the updated drivers, so I did, but then my screen would only run in 640x480, and the virus software started complaining, and MS Office wouldn't run!"
No kidding, this is what happened on my game system. It took several rounds of updates and yanking out my good video card and using a cheap one to get the machine bootable with the resulting mess.
OK, Star Trek transporters ws hyperbole, but not by much. Building a large-scale mass driver is still wildly infeasible: the switching of large magnetic fields at high propagation rates has turned on to be far more difficult than anticipated when mass drivers were first proposed. Like computers that understand spoken English, the proof of concept is easy. Getting it to work has turned out to be a bear.
That is not Rutan's fault. The difficulties he faced were not so much physical as political, with NASA's blatant and covert interference with his project. Take a look at the numbers. Rutan and his cohorts could replace the entire Shuttle operation in 2 years with half the annual budget, if NASA would fire the bureaucrats who are actively interfering with his project. The basic technologies of his design are more modern, simpler to implement, and a prime example of a "skunkworks" project of brillienat people with a good budget and excellent managers not interested in building fiefdoms at the expense of the project.
The short answer is "no". No one has ever built an effective mass driver: the technology is about as available as Star Trek transporters, and should be considered as likely to work until someone actually does the design and testing of even a small one for launching significant masses across a few hundred feet, much less launching to LEO.
Agreed about the corporate welfare payments, but don't forget the bureaucracy welfare payments to NASA's current management staff.
Dump NASA, or get a decent mandate in place to get out of the way of the Phoenix and its developers, and watch a Shuttle built the right way within 10 years. I'm wildly impressed by the Phoenix developers, they're doing all their basic engineering and most of their bureaucracy correctly.
Unfortunately, a lot of old systems don't have bootable USB accessible without yanking the system away from the wall and plugging in the back, or use weird chipsets that it's hard to build kernels to deal with.
The live bootable CD's and DVD's work very well because booting from the built-in CD or DVD is something any modern system absolutely must support. They're also very useful in that you can't write anything on them, and this helps prevent all sorts of potential security concerns, unlike the bootable USB devices.
Look, the OS really doesn't matter. What does matter is getting your employers to not do stupid things, like run their laptops without security patches and insist on running NFS and file sharing from home and on every machine in your group, getting them to pick decent passwords, teaching people never to use .zip attachments for anything, never running passphraseless accounts and open access points, etc., etc., etc.
Until you can get basic security steps like those in place, the world's best firewall is like a really big lock on a 3 foot high fence. Even the most casual crackers will simply step over it.
Install Linux on hardware that is decommissioned for running Windows, and on which you don't have the licenses to run Windows. Especially do this on your desktop machine to get the debugging tools and viral immunity, and use the remote desktop tools to log into Windows Pro boxes as necessary to administer them remotely. Then stop buying new Windows servers and make faster, more secure servers out of the decommissioned Windows boxes. And enjoy.
GOOD FOR YOU! Knowledgable, competent cross-platform work is a blessing and should be encouraged.
They see you as encroaching on their turf, rather than filling their needs by getting them the other resources they need. It drives the fiefdom leaders nuts when you point out how to save computers, money, and licenses by using superior tools that didn't come from their consultants or their "approved vendor", because it didn't come from their "process".
Establishing the trust needed to be given information, and to share ideas, in those other fields is vital and can be hard. So many outsiders to the field give such half-assed advice that people can set their filters very high to keep from having their time wasted. That's a human skill, and a precious one.
The future is the Trusted Computing Initiative: the activation codes will be node-locked to the BIOS or the CPU, with decently well encrypted keys, and you had better believe that these systems will be calling back to the mothership with your locally generated public keys to verify your ownership of your software, or even your ability to boot the machine because you've purchased an OS with a registered kernel.
Graceful key management without having to staple in tools like "keychains", a nice UI, really good VT100 and other terminal emulation that is vatly superior to that of the CygWin command prompt or the Windows telnet tools, good SSH tunneling management, and someone you can call at 3 AM with your software problem to walk you through debugging the issues. SecureCRT is a very sweet tool, and I recommend it.
Now, if you compared it to *Putty*, the open source OpenSSH based free tool built for Windows, you'd have a better basis for comparison.
If you're a piracy crook, you've got remote access to someone else's machine, possibly to hundreds of them. You build, or find, a pirate software website that publishes these.
The invested time is really not very large and scales nicely.
They also don't have the local bandwidth requirements of, say, a banking facility processing stock predictions and transactions that needs to pass many Gigabytes of data among a local cluster. Their bandwidth is more limited by their external access, and usage by the last mile of cable to their user's machines. Good hotswap and RAID5 is expensive, especially if you want to buy the good 3Ware or Adaptec stuff instead of that abomiable and undocumented Promise stuff you see in desktop motherboards.
They're all the same woman. It's amazing what you can do with a false nose and glasses....
He's not paying for his water, but I'm sure his landlord is and is about to go ballistic about this.