Oh, people still view Usenet. But if you do a fast traffic analysis, it's mostly for porn in the alt.binaries groups. Funny, since the flaw is in image viewing, this is exactly what's likely to carry such a worm or virus.
No, it's not. I've actually resigned from a job rather than engage in ludicrously unsafe practices, told by my boss to neglect security for a client that was in the contract in order to spend my time on software issues that were supposed to make the client forgive us for the deadlines we'd already missed. That didn't fix the practices, but it did get my name off them and let the client know I was unhappy about it.
Sometimes stupid or unsafe security practices are engaged in to save time or make a deadline, and especially to permit demoware to make a big sale. Demoware is death to secure computing, because what is a one-off is so often becomes a permanent feature.
RedHat and Novell have incorporated several of my modifications into their source code, with attributions, and I've contributed to perhaps a dozen different GPL products that RedHat and Novell use daily, and ancestors of products that they use as well. Microsoft ignores my patches and software bug reports on a weekly basis, with no meaningful response at all.
Trust me: programmers have grandmas, too. It's a lot of fun to go to the local computer store and help guide innocent people to the right solution, just for fun. It frees up the sales staff to answer my more detailed question, and they still get to put their little sales sticker on the product sold.
Oh, I've seen that done for open source work, too. I've actually ripped out and replaced some big chunks of very strange open source work that people had done with the hottest greatest thing since sliced bread, because it simply wasn't stable or up to the load.
Replacing a Qmail/LDAP server with Sendmail/NIS some years ago was the classic one, because while the new nifty stuff had some cute futures it simply wasn't ready to hold up under the load of the older software.
But Aunt Tillie doesn't want "control". She wants to do simple things in a simple way, and frankly as an engineer, so do I. Eric Raymond wrote about this problem compellingly in an old Slashdot referenced story, http://www.catb.org/~esr/writings/cups-horror.html.
But the big problem that contributes even more to an unhealthy and unhappy workplace is refusing to respond to legitimate issues and complaints caused by a specific problem employee, often the CEO, then censoring them as "whining" and "negative thinking" and using them to start a paper trail to dump that "unfaithful" employee who was actually trying to fix real problems.
I've actually seen this, when an engineer started doing a real hardware audit to bring deployed systems into the inventory and hardware management. It could have expanded the company's computer available hardware by 15% re-commissioning lost and "disabled" machines without spending money on new hardware, but it also turned out that it would have revealed the incompetence of the shipping manager, who had been losing machines and even stealing machines by sending them to his buddy and losing the receipts. The engineer got laid off with a new pair of twins to support and three weeks severance: two months later, when caught in the act, the shipping manager got an "oportunity to pursue other career options" with a six month severance package, his stock options intact, and polite references in order to avoid any press coverage of his theft.
Right. It's just what Linux needs, a bunch of kids who work longer hours, hold down 3 jobs, beat us on all our SAT's, and beat us at ping-pong. I loved that S. Korean kid at school, beating him at anything was harder than heck but worth every moment of the effort.
No need to hire a huge support staff. He can just use all of S. Kore as his new outsourced call center for the US, and with the money he doesn't pay them, he can build a real local support center in N. Korea with people who speak the local language.
No, they're lobbying to save themselves money and to get a uniform standard of privacy laws that will somehow, magically be consistent with the default options of their customer database services, and will be far more generous than many state policies.
As has been pointed out, look at the CAN-SPAM law and how it has actually shielded spam from state law and blocked no spam whatsoever. Then look at Microsoft's history of the "Sender-ID" email signature system, where all you have to do to spam past Microsoft's filters is buy a Sender-ID key from them. They will try to sell permission and tools to share the private data in exactly the way the states are trying to block right now. You can bet good money on it.
Sun's old dance with AT&T's SysV UNIX, back when Solaris was first released, has been in many ways a disaster. Basically, many open source tools have been carefully retrofitted to work on top of Solaris to provide access to BSD style tools and their more robust contemporary versions. Those open source tools also form the core of the Linux *operating system distributions*, not the Linux kernel itself.
So it's no surprise at all that Debian can be built on top of a Solaris kernel: few of the core OS packages care that much about the kernel. As long as you have the set of core header files describing the necessary hooks into the kernel, everything else basically falls into place. And those packages have worked with the Solaris since the beginning of Solaris.
But Sun still has a user community to work with. Their hardware is much more robust than PC hardware of similar vintage and cost, and their kernels deal very well with being really hammered by huge loads in a way that Linux kernels never have. Those people may appreciate a Debian variant for their uses, to ease their cross-platform support costs but provide the robust solutions that Suns have built for years.
Being a zealot does not make one illogical, anymore than being a Christian slave-owner makes one inconsistent. By choosing the rules to follow and defining them rigidly (such as the Bible's several clauses about how slaveowners should handle their slaves), one can easily follow those rules all the way to absurdity.
Similarly, being a zealot does not mean you're wrong. We have zealots for every good cause known to humanity, and by being so powerful in their beliefs and so consistent in their actions, they create examples for the rest of us. The danger is when the zealotry discards one's peers and compatriots as heretics: this is what happened when XFree86 discarded the CygWin community and patches submitted for Windows specific compilation, and when they refused to accept assistance from their users. They were discarded and replaced with Xorg almost universally, and the underlying project has continued.
Free as in speech software is really useful that way: we can fragment, split, or fork from the original and do useful work. OpenSSH got its start this way from the last open source version of SSH, and we've gotten good tools out of it. It's also why Dan Bernstein's "open source" but "you can only modify and publish if you get my explicit permission, written in blood on the hide of your firstborn" copyrights on his source code have led to the demise or disuse of his technically superior but copyright encumbered tools.
My betting money says that the lawyers for SCO are being funded the same means as SCO these days: "partnerships" brokered by Microsoft for their other, not very good but useful if Microsoft sweetens the deal, services.
No, "Trusted Computing" is not designed to prevent this. It is designed to *enforce* it. By having an appropriately signed application, required to access appropriately signed and controlled hardware such as your CD or DVD drive or appropriately encrypted files found on your CD, DVD, or downloaded files, it's designed to prevent you from accessing content in your files or on your systems without the signed license keys from the vendor.
So will the disk usage tools like "du", the file-system probe commands like "find", and many other tools. They're also aware of hard-links and handle them correctly. Re-writing the hundreds of utilities for manipulting the file system, like all the "zip" variants, to deal with symlinks is going to be painful. And getting it integrated reliably into SMB file-sharing for Microsoft, and for the rest of us for Samba is going to be a nightmare from hell.
Symlinks are dangerous: when you do a "cd symlink\.; cd.." in Windows, where are you going to wind up?
No one can indemnify you from delusional lawsuits, unless they have the spare money to fend it off or to succeed well enough to take back their court costs from the plaintiff. MySQL developers just don't have that kind of spare money.
Also note, SCO's lawsuit is draining their corporate sponsorship from Microsoft dry, and it's pretty clear from their income statements that they are being sponsored by Microsoft. Take a look over at http://groklaw.net/ for details. They don't have the resources to file another frivolous stock-pumping lawsuit like this one after they lose to IBM, becuase they're spending their core resources to pursue it. So if you're scared of SCO, I don't think you should worry about them after this: they haven't generated genuinely new or useful products in years.
So, they do get looked at least occasionally. How much energy do you want to waste on things that have been repeatedly demonstrated as mistaken, wishful thinking, placebo effect, or even just plain fraud?
Great. You've described the reasons why crackpots think science will never be able to understand them. Now, do the experiment. Name two subjects that science fears to study.
So you only re-do the results that don't come out they way you expect? Don't you see a potential source of statistical bias there, or of only doing the experiment until it gives you the result you want, then stopping? I know that's the normal way to do political opinion polls, but I'd like to hold scientific research up to a higher standard.
I don't know: it seems quite possible to recharge that remote control with properly applied olive oil. Some is used for good home Italian cooking, then you have the restRQSTthe bottle.
Actually computer power in space has turned out to be a real problem. Modern, smaller CPU's and RAM's are vulnerable to radiation damage, which can both flip bits at a startling rate and can permanently damage small electronics. The very shielding which might protect against such damage itself becomes radioactive over time. Large masses of very simple shielding, such as water, may help reduce the damage and be easily layered to dissipate the radioactive layers first, but the tanks themselves will become irradiated by cosmic radiation.
The problem was first well documented back in the 1960's and 1970's during the first launches, when portable computer power used far more robust core memories, and has never been really solved. Radiation-hardened electronics tend to be deliberately bulky old technologies, thus resistant to such radiation damage.
Don't leave out that DRM angle. By using the "Trusted Computing" model for content management, it's clear that Microsoft intends to force a set of hardware and software upgrades to access their "media" content. And doing their multi-media distribution in Windows Media and Windows managed tools means that open source tools will not be allowed to access them, blocked by the proprietary software and by the consequences of the DMCA laws against reverse engineering such techniques.
There's also the possibility of fully equipped operating systems on bootable large-capacity DVD's. The live CD and live DVD's, such as Knoppix at www.knoppix.org are wonderful tools that can help eliminate the market for Windows upgrades and support agreements on large numbers of Windows licenses at big sites by simply providing a fully equipped operating system, preferably freeware, on a local DVD. Buy the licenses if needed, build *one* DVD image, and install it on 500 machines rather than spending expensive Windows support manpower. Having much larger DVD's available, paid for by the money saved on buying hard drives, can allow these systems the power to replace fully equipped Windows systems without having to pick and choose to make it fit within 600 Meg for CD's, or 4.7 Gig for typical DVD's.
Not at all. I'm assuming that this is best used for laptop "sleep" or "hibernate" states. Waiting all that time for a reboot is really painful when you hop from meeting room to meeting room and need to either shut it down and reboot, or walk around with the lid open.
Server rooms with lots of blades, such as Hollywood render farms, would also have a big benefit. Coupled with LinuxBIOS to ease boot times and boot management, it could allow easy shut-down and system recovery for idle systems and a very serious long-term power savings during all the idle times for such machines.
Slashdot Mirror
Oh, people still view Usenet. But if you do a fast traffic analysis, it's mostly for porn in the alt.binaries groups. Funny, since the flaw is in image viewing, this is exactly what's likely to carry such a worm or virus.
No, it's not. I've actually resigned from a job rather than engage in ludicrously unsafe practices, told by my boss to neglect security for a client that was in the contract in order to spend my time on software issues that were supposed to make the client forgive us for the deadlines we'd already missed. That didn't fix the practices, but it did get my name off them and let the client know I was unhappy about it.
Sometimes stupid or unsafe security practices are engaged in to save time or make a deadline, and especially to permit demoware to make a big sale. Demoware is death to secure computing, because what is a one-off is so often becomes a permanent feature.
RedHat and Novell have incorporated several of my modifications into their source code, with attributions, and I've contributed to perhaps a dozen different GPL products that RedHat and Novell use daily, and ancestors of products that they use as well. Microsoft ignores my patches and software bug reports on a weekly basis, with no meaningful response at all.
Which of these is "supported"?
Trust me: programmers have grandmas, too. It's a lot of fun to go to the local computer store and help guide innocent people to the right solution, just for fun. It frees up the sales staff to answer my more detailed question, and they still get to put their little sales sticker on the product sold.
Oh, I've seen that done for open source work, too. I've actually ripped out and replaced some big chunks of very strange open source work that people had done with the hottest greatest thing since sliced bread, because it simply wasn't stable or up to the load. Replacing a Qmail/LDAP server with Sendmail/NIS some years ago was the classic one, because while the new nifty stuff had some cute futures it simply wasn't ready to hold up under the load of the older software.
But Aunt Tillie doesn't want "control". She wants to do simple things in a simple way, and frankly as an engineer, so do I. Eric Raymond wrote about this problem compellingly in an old Slashdot referenced story, http://www.catb.org/~esr/writings/cups-horror.html .
But the big problem that contributes even more to an unhealthy and unhappy workplace is refusing to respond to legitimate issues and complaints caused by a specific problem employee, often the CEO, then censoring them as "whining" and "negative thinking" and using them to start a paper trail to dump that "unfaithful" employee who was actually trying to fix real problems.
I've actually seen this, when an engineer started doing a real hardware audit to bring deployed systems into the inventory and hardware management. It could have expanded the company's computer available hardware by 15% re-commissioning lost and "disabled" machines without spending money on new hardware, but it also turned out that it would have revealed the incompetence of the shipping manager, who had been losing machines and even stealing machines by sending them to his buddy and losing the receipts. The engineer got laid off with a new pair of twins to support and three weeks severance: two months later, when caught in the act, the shipping manager got an "oportunity to pursue other career options" with a six month severance package, his stock options intact, and polite references in order to avoid any press coverage of his theft.
Right. It's just what Linux needs, a bunch of kids who work longer hours, hold down 3 jobs, beat us on all our SAT's, and beat us at ping-pong. I loved that S. Korean kid at school, beating him at anything was harder than heck but worth every moment of the effort.
No need to hire a huge support staff. He can just use all of S. Kore as his new outsourced call center for the US, and with the money he doesn't pay them, he can build a real local support center in N. Korea with people who speak the local language.
And oh, yes, add the obligatory:
4: Profit!!!!
No, they're lobbying to save themselves money and to get a uniform standard of privacy laws that will somehow, magically be consistent with the default options of their customer database services, and will be far more generous than many state policies.
As has been pointed out, look at the CAN-SPAM law and how it has actually shielded spam from state law and blocked no spam whatsoever. Then look at Microsoft's history of the "Sender-ID" email signature system, where all you have to do to spam past Microsoft's filters is buy a Sender-ID key from them. They will try to sell permission and tools to share the private data in exactly the way the states are trying to block right now. You can bet good money on it.
Sun's old dance with AT&T's SysV UNIX, back when Solaris was first released, has been in many ways a disaster. Basically, many open source tools have been carefully retrofitted to work on top of Solaris to provide access to BSD style tools and their more robust contemporary versions. Those open source tools also form the core of the Linux *operating system distributions*, not the Linux kernel itself.
So it's no surprise at all that Debian can be built on top of a Solaris kernel: few of the core OS packages care that much about the kernel. As long as you have the set of core header files describing the necessary hooks into the kernel, everything else basically falls into place. And those packages have worked with the Solaris since the beginning of Solaris.
But Sun still has a user community to work with. Their hardware is much more robust than PC hardware of similar vintage and cost, and their kernels deal very well with being really hammered by huge loads in a way that Linux kernels never have. Those people may appreciate a Debian variant for their uses, to ease their cross-platform support costs but provide the robust solutions that Suns have built for years.
Being a zealot does not make one illogical, anymore than being a Christian slave-owner makes one inconsistent. By choosing the rules to follow and defining them rigidly (such as the Bible's several clauses about how slaveowners should handle their slaves), one can easily follow those rules all the way to absurdity.
Similarly, being a zealot does not mean you're wrong. We have zealots for every good cause known to humanity, and by being so powerful in their beliefs and so consistent in their actions, they create examples for the rest of us. The danger is when the zealotry discards one's peers and compatriots as heretics: this is what happened when XFree86 discarded the CygWin community and patches submitted for Windows specific compilation, and when they refused to accept assistance from their users. They were discarded and replaced with Xorg almost universally, and the underlying project has continued.
Free as in speech software is really useful that way: we can fragment, split, or fork from the original and do useful work. OpenSSH got its start this way from the last open source version of SSH, and we've gotten good tools out of it. It's also why Dan Bernstein's "open source" but "you can only modify and publish if you get my explicit permission, written in blood on the hide of your firstborn" copyrights on his source code have led to the demise or disuse of his technically superior but copyright encumbered tools.
My betting money says that the lawyers for SCO are being funded the same means as SCO these days: "partnerships" brokered by Microsoft for their other, not very good but useful if Microsoft sweetens the deal, services.
No, "Trusted Computing" is not designed to prevent this. It is designed to *enforce* it. By having an appropriately signed application, required to access appropriately signed and controlled hardware such as your CD or DVD drive or appropriately encrypted files found on your CD, DVD, or downloaded files, it's designed to prevent you from accessing content in your files or on your systems without the signed license keys from the vendor.
I see that I have been trolled, fairly successfully.
So will the disk usage tools like "du", the file-system probe commands like "find", and many other tools. They're also aware of hard-links and handle them correctly. Re-writing the hundreds of utilities for manipulting the file system, like all the "zip" variants, to deal with symlinks is going to be painful. And getting it integrated reliably into SMB file-sharing for Microsoft, and for the rest of us for Samba is going to be a nightmare from hell.
.." in Windows, where are you going to wind up?
Symlinks are dangerous: when you do a "cd symlink\.; cd
No one can indemnify you from delusional lawsuits, unless they have the spare money to fend it off or to succeed well enough to take back their court costs from the plaintiff. MySQL developers just don't have that kind of spare money.
Also note, SCO's lawsuit is draining their corporate sponsorship from Microsoft dry, and it's pretty clear from their income statements that they are being sponsored by Microsoft. Take a look over at http://groklaw.net/ for details. They don't have the resources to file another frivolous stock-pumping lawsuit like this one after they lose to IBM, becuase they're spending their core resources to pursue it. So if you're scared of SCO, I don't think you should worry about them after this: they haven't generated genuinely new or useful products in years.
So, they do get looked at least occasionally. How much energy do you want to waste on things that have been repeatedly demonstrated as mistaken, wishful thinking, placebo effect, or even just plain fraud?
Great. You've described the reasons why crackpots think science will never be able to understand them. Now, do the experiment. Name two subjects that science fears to study.
So you only re-do the results that don't come out they way you expect? Don't you see a potential source of statistical bias there, or of only doing the experiment until it gives you the result you want, then stopping? I know that's the normal way to do political opinion polls, but I'd like to hold scientific research up to a higher standard.
You wrote: "you'd have to figure out how to attatch electrodes behind the lobes of someone's ears at range".
What do you think those Ipod headphones are really for?
I don't know: it seems quite possible to recharge that remote control with properly applied olive oil. Some is used for good home Italian cooking, then you have the restRQSTthe bottle.
Actually computer power in space has turned out to be a real problem. Modern, smaller CPU's and RAM's are vulnerable to radiation damage, which can both flip bits at a startling rate and can permanently damage small electronics. The very shielding which might protect against such damage itself becomes radioactive over time. Large masses of very simple shielding, such as water, may help reduce the damage and be easily layered to dissipate the radioactive layers first, but the tanks themselves will become irradiated by cosmic radiation.
The problem was first well documented back in the 1960's and 1970's during the first launches, when portable computer power used far more robust core memories, and has never been really solved. Radiation-hardened electronics tend to be deliberately bulky old technologies, thus resistant to such radiation damage.
Don't leave out that DRM angle. By using the "Trusted Computing" model for content management, it's clear that Microsoft intends to force a set of hardware and software upgrades to access their "media" content. And doing their multi-media distribution in Windows Media and Windows managed tools means that open source tools will not be allowed to access them, blocked by the proprietary software and by the consequences of the DMCA laws against reverse engineering such techniques.
There's also the possibility of fully equipped operating systems on bootable large-capacity DVD's. The live CD and live DVD's, such as Knoppix at www.knoppix.org are wonderful tools that can help eliminate the market for Windows upgrades and support agreements on large numbers of Windows licenses at big sites by simply providing a fully equipped operating system, preferably freeware, on a local DVD. Buy the licenses if needed, build *one* DVD image, and install it on 500 machines rather than spending expensive Windows support manpower. Having much larger DVD's available, paid for by the money saved on buying hard drives, can allow these systems the power to replace fully equipped Windows systems without having to pick and choose to make it fit within 600 Meg for CD's, or 4.7 Gig for typical DVD's.
Not at all. I'm assuming that this is best used for laptop "sleep" or "hibernate" states. Waiting all that time for a reboot is really painful when you hop from meeting room to meeting room and need to either shut it down and reboot, or walk around with the lid open.
Server rooms with lots of blades, such as Hollywood render farms, would also have a big benefit. Coupled with LinuxBIOS to ease boot times and boot management, it could allow easy shut-down and system recovery for idle systems and a very serious long-term power savings during all the idle times for such machines.