Yes, but it's hard to predict what patches require reboots and what patches will subtlely change certain services. And Windows Software Update Services normally isn't applied to personal machines and personal laptops and production services managed by third parties: scheduling one day a month to risk important services and make sure your IT staff is ready to stay late and roll things back out if they break is pretty important.
Wow. I haven't had to do that kind of debugging in years: you have my respect, sir.
But wouldn't it have been nicer to run it in gdb with the source code, or in ddd, and been able to find the broken source and patch it for the future?
And as "Trusted Computing" takes off, and starts doing cryptographic signing of system binaries like MS-Office and core system files like Internet Explorer's dll's, isn't your patch going to be detected as a gross security violation and cause the security tools to start shrieking about it? Or will you be able to get the signatures needed to authenticate your changes?
Sir, please hop over to http://www.groklaw.net/ and take a look at exactly what SCO has beein trying to subpoena for the "discovery" stage of their lawsuit with IBM. It is literally millions of lines of code, entire changelog histories, hundreds of hours of testimony, and other vast amounts of evidence.
Their limited success getting access to this material means that it is at least thinkable to grant such wide access to the material. Fortunately, IBM's lawyers have actually been doing their jobs in defending against such wide-ranging, expensive, and business plan revealing fishing expeditions.
It's not impossible for a Jew to plot to overthrow world governments, either.
But when you hear the company that stole VMS to create Windows NT, that stole the design for the Microsoft mouse, and that regularly gets caught stealing ideas from partners, and that is funding the SCO lawsuite against IBM and after years of nasty litigation can't find a single line of stolen code, and when you see the clear history of who contributed what lines of code to what tools in Linux's changelogs so that there's complete transparency of who brought in what in the Linux world, you see this as part of yet another Microsoft FUD campaign.
If Vista's unavailable, maybe XBox-360 based systems, and use the hull as the heat sink. You'll only be able to use it in the Arctic Circle with all systems active, though.
I've dealt with Dell from a number of businesses, large and small. Their customer support has always been good: not as good as the guy around the corner who owes me a lot of beer, sometimes, but always several steps up from their competitors. This is true even as they outsource more of it: they seem to be giving the "My name is Mahareesh, but you can call me Jack" first tier tech support enough authority to actually replace parts rather than forcing me through all the "have you done this Windows re-install trick?" on my Linux boxes.
I don't know where you're getting your Dell experience form, but they've been well worth the premium some folks pay for them in a production environment.
No joke. The spaghetti code and "make the ship date, even if it's vulnerable" approach has finally come home to roost. It's one of the reasons Vista is taking so long: they're actually forcing their own people to do the QA up front, and having to re-write a bunch of old rehashed DOS and Win9x tools from scratch because they're so awful.
Innovation is fine as a mission, but the market won't stand much more "innovation" that breaks important things, and Microsoft seems to be taking it seriously for Vista. That doesn't mean that Vista will be secure, or complete, but there are people there trying and getting some management support they didn't get a few years ago.
Linux hardly came from nothing: Linux, the kernel, was the lynchpin finally completing many years of devoted work by the Free Software Foundation and their collaborators creating the gcc compiler and Emacs, by MIT developing X11, and since then by lots of people creating tools because they needed the job done, not because they were looking to sell off the software itself. It's also based in basic style on UNIX, especially the filesystem and command line tools (which were re-written in open source style by the Free Software Foundation).
That turned out to be effort well-spent and well-invested by the developers: but it doesn't fit ordinary business plans, and wouldn't have happened in the Microsoft business models of "lock out competitors".
They do know. Microsoft has apparently done a major re-engineering of their entire software production process, with a "fix it *before* we run the test" and an actual "regression test the whole shmear" process. Both have huge costs in the short term, becuase a lot of stuff that would be swept under the carpet and "fixed later" such as the blatant security violations of Microsoft tools playing fast and loose with what should be privileged operations are getting shot down early, and people are being forced to do it right.
In the short term, that costs a lot of man-hours, especially as brilliant and fast but erratic programmers have to relearn how to do things or have their errors thrown up in their face this quarter, rather than 3 years down the line when the next nasty worm takes down SQL servers all over the world due to a stupid default password setting and wide open access by default. Re-engineering that large a company takes time: they do seem to be taking the process seriously, but of course it's going to slow product releases planned by managers who didn't expect to have to write good code this quarter.
There are real advantages for the consumer: getting code that is release quality, or at least late beta quality instead of using the customer base as a bunch of alpha testers saves the average customer a huge amount of work and expense. An application that arrives months later but works out of the box is more valuable than a broken one that can't operate until the patches happen that much time later.
The delays to Vista are hurting that model. Windows XP has matured and become reasonably stable. Windows 2000 is falling off the support wagon, but with the various anti-virus and anti-spyware tools currently available, the upgrade to Vista cannot be justified by its limited built-ni anti-spyware capabilities.
Remember, Microsoft is a big believer in letting users activate and install stupid things by just clicking on them. It's a big part of their sales demos, so people are likely to continue to click on "cool screensavers" and other nonsense and rely on their anti-spyware tool to block stuff that is, in fact spyware but which they've just unknowingly authorized for installation. And Microsoft, as a big corporate entity, dare not judge spyware fiercely or they will be facing lawsuits from a bunch of companies that will drain money they don't want to spend.
The important words in your note are "cash out". The same scumbags who brought us Gator are now part of Microsoft, and as such constitute a legal risk to Microsoft's very deep pockets. So it makes complete sense for all the VP's and stock-bearing staff to cash out immediately and abandon the company to whatever use Microsoft might have for their office space and their customer database management tools.
There are some other tools and approaches Microsoft might find technically useful from Gator, such as the tendency to insinuate itself into the OS in ways that are incredibly painful to flush. That expertise is deep and expensive to develop, and could be useful to Microsoft development teams that want to weave in things like Windows Media Player, the Microsoft search tools to displace the Google toolbar, etc.
China is still a massive source of spam and spyware. It's also a stunningly huge site for pirated software, especially Windows, Office, games, and DVD movies. The Chinese government takes no action again anything that brings in any revenue or avoids expenses for its populace, but anything that threatens the existing power structure (such as free speech or encrypted email) are slapped down hard.
Spyware may be illegal there, but they're doing even less about it than the US government.
You've got it. Since Microsoft bought them, they need to avoid poisoning the well for more important Microsoft products like Internet Explorer and Windows by pushing adware that helps bog down Windows systems and show its lack of security and softwrae management.
What I'd like to see out of Claria is a published list of the authors of the software and most especially of their manager, to use as an employment blacklist for more ethical companies.
The differential is even worse on a drive that is fragmented: those 8 blocks may be scattered all over the disk and require additional fascinating head movements to get them all off in order.
There are losses of file space for file systems that contain lots of small files: the maximum space wasted by having only one byte on a block goes from 511 bytes to 4095 bytes, and that will affect available disk space in some systems that use lots of small files.
A file-system block is not a hard-disk block. This means that block sizes smaller than 4096 bytes will not be available, and that tools that talk to the disk at a low level (such as fdisk and parted) will have to be reviewed for any assumptions that block sizes are not, in fact, 512 bytes. It also means that old drivers that made such assumptions are not going to interoperate correctly with these new disks and controllers, unless the manufacturers are very clever about maintaining interfaces that look identical.
I expect a lot of people running old OS's who don't like to upgrade are going to be very unhappy when they can't use the new hardware and tell their poor IT person to "just find the patch!" when they can't transfer their deprecated OS and tools to the new "our sales partner recommended it!" hardware. Vista should be OK with it: Microsoft has strong partnerships with the manufacturers and the drives absolutely must work under Vista. But the NetBSD authors and the people running RedHat 6.x servers because "we know our code works on it" are in for a big surprise if this becomes popular.
Expect to see it first on the very large hard drives, 200 Gigabytes and up, where the larger block size is a real advantage.
Most Spyware or Adware vendors and their programmers are frankly incompetent. They grow in number because the lure of "Making Money on the Information $uperhighway" is so large that despite the failures of most of them, they can find new idiots to join their legion.
There are a very few really technically competent spyware and adware programmers: they sell their tools to the wanna-be's, and to commercial companies that make stupid decisions based on hype, to stay in business. Witness the Sony DRM fiasco: what could have been a well-written DRM tool was made far too powerful and aggressive. The original installation was reasonably clever and competent: it just reached way too far into the operating system to protect itself and to take control of the user's computer, instead of just protecting itself.
Are you under the impression that the authors of that piece of malware didn't get paid? Or that they didn't change their company name and go on to write other DRMware for other companies?
I'm sorry to say, but Microsoft Vista may hurt some cheesy anti-spyware and adware makers out of business. It may even put some spyware and adware creators out of business, which is much better to do. But the fundamental "features are more important than security" and "exciting demos that any idiot can run are more important than robust code that follows the API" approaches that are fundamental to Microsoft software will continue to create massive loopholes for adware and spyware.
Remember, a lot of adware and spyware and DRMware and other unwanted tools are being included with downloads, installed from vendor's CD's, and causing endless trouble for innocent users who didn't know it was there. This vulnerability isn't going away: users will continue to click on the "Yes, I will let the angel of death come and take my firstborn, and I promise not to mark my door mantel to keep him out" user agreements without reading them. Anti-spyware tools will continue to be sold to flush web cookies used to track your browsing and purchasing, dangerous DRMware like the Sony rootkit from the Sony audio CD's, and other tools that the anti-virus makers are very reluctant to touch.
Simply improving Vista's security models over those of Windows leaves a lot of growth room for these manipulative vendors of "speed up your web browsing 400%" tools.
I believe you're referring to the recent theft of technologies for tantalum bypass capacitors: unfortunately, the technology thief didn't get all the details of the technology, and the result was a lot of failure-prone motherboards coming out of Taiwan and failing after less than a year in the field. Dell, in particular, had a problem with them but was very good about replacing the motherboards at the first sign of trouble. They got some good support reviews for dealing so well with something clearly not their fault.
As I remember, when it came out, there were patent issues with RSA in the PGP encryption. RSA owned the patents on RSA encryption used by PGP, but refused to release them for other uses, even if you phoned them up and tried to pay them for its use in another open source project. (I know: I tried.) RSA seemed more concerned aobut its inclusion in exported software than in actually making the software available. There were real USA export regulations about it, classifying encryption as a war material. Those regulations were found unconstitutional, and transferred from Customs to Commerce instead of thrown out, and we're still seeing them in the courts occasionally. It's why US users had such trouble using SSL keys as long as 128 bits on so many web browsers for such a long time.
So PGPphone had a serious legal battle to reach wide use, and it never caught on as well as not having a great user interface. I hope Phil's newer tool catches on.
Slashdot Mirror
Yes, but it's hard to predict what patches require reboots and what patches will subtlely change certain services. And Windows Software Update Services normally isn't applied to personal machines and personal laptops and production services managed by third parties: scheduling one day a month to risk important services and make sure your IT staff is ready to stay late and roll things back out if they break is pretty important.
Wow. I haven't had to do that kind of debugging in years: you have my respect, sir.
But wouldn't it have been nicer to run it in gdb with the source code, or in ddd, and been able to find the broken source and patch it for the future?
And as "Trusted Computing" takes off, and starts doing cryptographic signing of system binaries like MS-Office and core system files like Internet Explorer's dll's, isn't your patch going to be detected as a gross security violation and cause the security tools to start shrieking about it? Or will you be able to get the signatures needed to authenticate your changes?
Sir, please hop over to http://www.groklaw.net/ and take a look at exactly what SCO has beein trying to subpoena for the "discovery" stage of their lawsuit with IBM. It is literally millions of lines of code, entire changelog histories, hundreds of hours of testimony, and other vast amounts of evidence.
Their limited success getting access to this material means that it is at least thinkable to grant such wide access to the material. Fortunately, IBM's lawyers have actually been doing their jobs in defending against such wide-ranging, expensive, and business plan revealing fishing expeditions.
You left out this part -steal VMS kernel from DEC -slap Windows like interface on top -sell as WinNT
OK, I'm invoking Godwin's Law.
It's not impossible for a Jew to plot to overthrow world governments, either.
But when you hear the company that stole VMS to create Windows NT, that stole the design for the Microsoft mouse, and that regularly gets caught stealing ideas from partners, and that is funding the SCO lawsuite against IBM and after years of nasty litigation can't find a single line of stolen code, and when you see the clear history of who contributed what lines of code to what tools in Linux's changelogs so that there's complete transparency of who brought in what in the Linux world, you see this as part of yet another Microsoft FUD campaign.
If Vista's unavailable, maybe XBox-360 based systems, and use the hull as the heat sink. You'll only be able to use it in the Arctic Circle with all systems active, though.
That's why you have the tapes from every date: so you can look things up later in case you wind up in court.
I've dealt with Dell from a number of businesses, large and small. Their customer support has always been good: not as good as the guy around the corner who owes me a lot of beer, sometimes, but always several steps up from their competitors. This is true even as they outsource more of it: they seem to be giving the "My name is Mahareesh, but you can call me Jack" first tier tech support enough authority to actually replace parts rather than forcing me through all the "have you done this Windows re-install trick?" on my Linux boxes.
I don't know where you're getting your Dell experience form, but they've been well worth the premium some folks pay for them in a production environment.
No joke. The spaghetti code and "make the ship date, even if it's vulnerable" approach has finally come home to roost. It's one of the reasons Vista is taking so long: they're actually forcing their own people to do the QA up front, and having to re-write a bunch of old rehashed DOS and Win9x tools from scratch because they're so awful.
Innovation is fine as a mission, but the market won't stand much more "innovation" that breaks important things, and Microsoft seems to be taking it seriously for Vista. That doesn't mean that Vista will be secure, or complete, but there are people there trying and getting some management support they didn't get a few years ago.
Linux hardly came from nothing: Linux, the kernel, was the lynchpin finally completing many years of devoted work by the Free Software Foundation and their collaborators creating the gcc compiler and Emacs, by MIT developing X11, and since then by lots of people creating tools because they needed the job done, not because they were looking to sell off the software itself. It's also based in basic style on UNIX, especially the filesystem and command line tools (which were re-written in open source style by the Free Software Foundation).
That turned out to be effort well-spent and well-invested by the developers: but it doesn't fit ordinary business plans, and wouldn't have happened in the Microsoft business models of "lock out competitors".
They do know. Microsoft has apparently done a major re-engineering of their entire software production process, with a "fix it *before* we run the test" and an actual "regression test the whole shmear" process. Both have huge costs in the short term, becuase a lot of stuff that would be swept under the carpet and "fixed later" such as the blatant security violations of Microsoft tools playing fast and loose with what should be privileged operations are getting shot down early, and people are being forced to do it right.
In the short term, that costs a lot of man-hours, especially as brilliant and fast but erratic programmers have to relearn how to do things or have their errors thrown up in their face this quarter, rather than 3 years down the line when the next nasty worm takes down SQL servers all over the world due to a stupid default password setting and wide open access by default. Re-engineering that large a company takes time: they do seem to be taking the process seriously, but of course it's going to slow product releases planned by managers who didn't expect to have to write good code this quarter.
There are real advantages for the consumer: getting code that is release quality, or at least late beta quality instead of using the customer base as a bunch of alpha testers saves the average customer a huge amount of work and expense. An application that arrives months later but works out of the box is more valuable than a broken one that can't operate until the patches happen that much time later.
If Ballmer could hit his targets, he wouldn't be in this fix.
The delays to Vista are hurting that model. Windows XP has matured and become reasonably stable. Windows 2000 is falling off the support wagon, but with the various anti-virus and anti-spyware tools currently available, the upgrade to Vista cannot be justified by its limited built-ni anti-spyware capabilities.
Remember, Microsoft is a big believer in letting users activate and install stupid things by just clicking on them. It's a big part of their sales demos, so people are likely to continue to click on "cool screensavers" and other nonsense and rely on their anti-spyware tool to block stuff that is, in fact spyware but which they've just unknowingly authorized for installation. And Microsoft, as a big corporate entity, dare not judge spyware fiercely or they will be facing lawsuits from a bunch of companies that will drain money they don't want to spend.
The important words in your note are "cash out". The same scumbags who brought us Gator are now part of Microsoft, and as such constitute a legal risk to Microsoft's very deep pockets. So it makes complete sense for all the VP's and stock-bearing staff to cash out immediately and abandon the company to whatever use Microsoft might have for their office space and their customer database management tools.
There are some other tools and approaches Microsoft might find technically useful from Gator, such as the tendency to insinuate itself into the OS in ways that are incredibly painful to flush. That expertise is deep and expensive to develop, and could be useful to Microsoft development teams that want to weave in things like Windows Media Player, the Microsoft search tools to displace the Google toolbar, etc.
China is still a massive source of spam and spyware. It's also a stunningly huge site for pirated software, especially Windows, Office, games, and DVD movies. The Chinese government takes no action again anything that brings in any revenue or avoids expenses for its populace, but anything that threatens the existing power structure (such as free speech or encrypted email) are slapped down hard.
Spyware may be illegal there, but they're doing even less about it than the US government.
You've got it. Since Microsoft bought them, they need to avoid poisoning the well for more important Microsoft products like Internet Explorer and Windows by pushing adware that helps bog down Windows systems and show its lack of security and softwrae management.
What I'd like to see out of Claria is a published list of the authors of the software and most especially of their manager, to use as an employment blacklist for more ethical companies.
Emacs. Let the vi/Emacs flame-wars begin!
200 is still pretty generous for low end systems, SCSI, and high speed arrays. But you have a point. Feel free to up it by a factor of 2 or more.
The differential is even worse on a drive that is fragmented: those 8 blocks may be scattered all over the disk and require additional fascinating head movements to get them all off in order.
There are losses of file space for file systems that contain lots of small files: the maximum space wasted by having only one byte on a block goes from 511 bytes to 4095 bytes, and that will affect available disk space in some systems that use lots of small files.
A file-system block is not a hard-disk block. This means that block sizes smaller than 4096 bytes will not be available, and that tools that talk to the disk at a low level (such as fdisk and parted) will have to be reviewed for any assumptions that block sizes are not, in fact, 512 bytes. It also means that old drivers that made such assumptions are not going to interoperate correctly with these new disks and controllers, unless the manufacturers are very clever about maintaining interfaces that look identical.
I expect a lot of people running old OS's who don't like to upgrade are going to be very unhappy when they can't use the new hardware and tell their poor IT person to "just find the patch!" when they can't transfer their deprecated OS and tools to the new "our sales partner recommended it!" hardware. Vista should be OK with it: Microsoft has strong partnerships with the manufacturers and the drives absolutely must work under Vista. But the NetBSD authors and the people running RedHat 6.x servers because "we know our code works on it" are in for a big surprise if this becomes popular.
Expect to see it first on the very large hard drives, 200 Gigabytes and up, where the larger block size is a real advantage.
Most Spyware or Adware vendors and their programmers are frankly incompetent. They grow in number because the lure of "Making Money on the Information $uperhighway" is so large that despite the failures of most of them, they can find new idiots to join their legion.
There are a very few really technically competent spyware and adware programmers: they sell their tools to the wanna-be's, and to commercial companies that make stupid decisions based on hype, to stay in business. Witness the Sony DRM fiasco: what could have been a well-written DRM tool was made far too powerful and aggressive. The original installation was reasonably clever and competent: it just reached way too far into the operating system to protect itself and to take control of the user's computer, instead of just protecting itself.
Are you under the impression that the authors of that piece of malware didn't get paid? Or that they didn't change their company name and go on to write other DRMware for other companies?
I'm sorry to say, but Microsoft Vista may hurt some cheesy anti-spyware and adware makers out of business. It may even put some spyware and adware creators out of business, which is much better to do. But the fundamental "features are more important than security" and "exciting demos that any idiot can run are more important than robust code that follows the API" approaches that are fundamental to Microsoft software will continue to create massive loopholes for adware and spyware.
Remember, a lot of adware and spyware and DRMware and other unwanted tools are being included with downloads, installed from vendor's CD's, and causing endless trouble for innocent users who didn't know it was there. This vulnerability isn't going away: users will continue to click on the "Yes, I will let the angel of death come and take my firstborn, and I promise not to mark my door mantel to keep him out" user agreements without reading them. Anti-spyware tools will continue to be sold to flush web cookies used to track your browsing and purchasing, dangerous DRMware like the Sony rootkit from the Sony audio CD's, and other tools that the anti-virus makers are very reluctant to touch.
Simply improving Vista's security models over those of Windows leaves a lot of growth room for these manipulative vendors of "speed up your web browsing 400%" tools.
I believe you're referring to the recent theft of technologies for tantalum bypass capacitors: unfortunately, the technology thief didn't get all the details of the technology, and the result was a lot of failure-prone motherboards coming out of Taiwan and failing after less than a year in the field. Dell, in particular, had a problem with them but was very good about replacing the motherboards at the first sign of trouble. They got some good support reviews for dealing so well with something clearly not their fault.
Would you mind telling us the hosting site's name? We'd like to install our password sniffers early. Or should we just monitor the FTP sites instead?
As I remember, when it came out, there were patent issues with RSA in the PGP encryption. RSA owned the patents on RSA encryption used by PGP, but refused to release them for other uses, even if you phoned them up and tried to pay them for its use in another open source project. (I know: I tried.) RSA seemed more concerned aobut its inclusion in exported software than in actually making the software available. There were real USA export regulations about it, classifying encryption as a war material. Those regulations were found unconstitutional, and transferred from Customs to Commerce instead of thrown out, and we're still seeing them in the courts occasionally. It's why US users had such trouble using SSL keys as long as 128 bits on so many web browsers for such a long time.
So PGPphone had a serious legal battle to reach wide use, and it never caught on as well as not having a great user interface. I hope Phil's newer tool catches on.