Microsoft designed a single user OS with no in built security in a time where networks were rare and have been forced to continue on with it by their customer base. All security ended up being tacked on because MS cant afford to kill legacy applications. I really don't think anyone at MS wants Windows to be insecure, it just happened that way and now they have to live with it.
Windows NT was designed from day one to be a multiuser OS. It's security is in no way "tacked on". The last version of Windows that fit your description was released a decade ago, and retired only a few years after that.
However, that's not the point. An AV is needed in Windows for the same reason it's needed on any other platform - it's a last ditch attempt at saving the end user when they've already circumvented the OS's security mechanisms. AV exists pretty much solely to protect the user from situations that OS security simply cannot.
A lot of commercial vendors treat independent researchers with contempt (how dare they find holes in our products) or as slaves (they should do the work our quality control dept should, for free)...
Of course, the folks who find a problem and then say "you have a week to fix this and then we release it into the wild" don't win their side any favours, either...
You didn't have to do this with Slackware in 1994.
Yes you did. Linux didn't get kernel modules until around 1995 (and they weren't in common usage until years after that).
Windows 7 doesn't even do this with hardware slightly older than it.
Windows 7 installs perfectly on my old Dell Precision M60 laptop, released in 2003. Not only that, but it actually works properly with my docking station and external monitors - something none of the popular Linux distributions manage to do.
This is probably a reference to this clause, found both in Vista and Windows 7 eulas. Available here in various eulas, such as Vista Home Premium English and Windows 7 Home Premium English, found here: microsoft Searching the XP sp2 eula does not seem to contain a similar clause.
So in other words, it bears only the vaguest and most indirect relevance to the OP's accusation. My, how surprising.
See their Trusted Platform Modules (TPM), DRM and HDCP divisions and reconsider our imaginary "freedom to run what we want" a double standard.
None of those do anything unless the application and/or user tells them to.
From Vista on a clause in every Windows EULA gives MS the right to delete executables and files from YOUR computer should they choose to do so (does XP also have this?)
Bullshit. IE has never been superior to any other browser.
The industry disagrees. IE4 was rapidly stealing marketshare from Navigator long before it was integrated into Windows 98. Heck, the _beta_ version was winning non-trivial numbers of people away from Navigator 3 (and the freshly released 4).
As for the unholy disaster that was Navigator 4.0, it's hard to imagine anyone saying it was better than _anything_ with a straight face, let alone IE4. They'd nearly made it tolerable by 4.5 or so, but by then it was pretty much game over.
For that matter Windows has never been superior to any other OS. I can't think of a single product Microsoft has ever made where I would label it "the best" product on the market.
Millions of people clearly feel Windows is the best option for them. I certainly do, and it has nothing to do with lack of knowledge or experience with the alternatives.
Um, when Microsoft made that claim, they were referring to Windows XP and Internet Explorer 6 which are both almost 9 years old.
Actually it was about Windows 98 and IE4.
At that time, IE6 was very likely tightly linked to the OS.
It was no more "linked" then than it is now.
The argument is, and has always been, that the integration of IE into Windows resulted in various other parts of the system being dependent on the shared IE components to work. This statement is 100% true, and the subsequent emulation of that design by basically every other major platform has long since vindicated it from a technical perspective.
The only time the "IE is part of the OS" claim is incorrect, is when people deliberately choose a poor context in which to (mis)interpret it. IE is "part of the OS" the same way Explorer, Control Panel, the network stack, the Win32 API, or any number of other components are part of the OS. It was true then, it's true now, and it's also true of all the other major platforms as well (OSX w/Safari+Webkit, KDE w/Konquerer+khtml, GNOME w/Epiphany+Webkit).
But a machine's administrator can easily replace/usr/bin/perl with any binary that implements a compatible interface.
As you could with IE.
Any/usr/bin/perl that parses and runs the same language will do, even if it has third-party defect fixes applied to it.
Unless, of course, there are programs that depend on some buggy behaviour a third-party replacement doesn't have.
Microsoft doesn't make that so easy with mshtml.dll: either you use Microsoft's mshtml.dll, or you don't use Windows.
Whether they make it easy or not is irrelevant (thought I fail to see how they make it "hard"). The principle - and therefore the argument - is the same.
That's because it was pretty much the only browser you had to "buy".
IE was superior to Navigator in terms of function and performance from IE4 onwards, primarily because Navigator was an utter disaster from version 4.0 onwards. Even comparing IE 3.x and Navigator 3.x, the difference was so insignificant as to be purely a matter of preference or inertia.
Netscape lost because their product sucked. No other reason of consequence.
Each kid born, and each kid nursed, reduces the risk of breast cancer. It's a 5% drop and a 7% drop, or the other way around. Assuming every kid is nursed, that should be about 12% to 13% risk reduction per kid.
Pity about all those other life-threatening complications that can occur during pregnancy.
A woman naturally has about a dozen kids. Breast cancer is quite rare in countries where women birth early and often.
And what's the average life expectancy of women in those countries ?
What is the basis of your assumption that it must serve a purpose ?
Homosexuality is likely a simple random mutation, which is why it shows up at roughly the same rate regardless of other factors, and why that rate doesn't change over time.
The value of marriage, compared to homosexuality, is that it usally leads to children, which will grow up and then pay taxes that are necessary sustain our standards of living in the long term.
Which is irrelevant, since homosexuals being married doesn't prevent any of that.
That's the basis for all pro-family / pro-marriage / etc. policies.
They're not "pro-marriage", they're "anti-gay". If they were just "pro-marriage", then they wouldn't give a damn about the gender of the two people getting married, since the likelihood of a married and unmarried gay couple producing children is the same, and has zero impact on whether or not a heterosexual couple will produce children (within marriage or otherwise).
It's not a moral debate, just a social one.
It's absolutely a moral debate. There's no social reason to prevent homosexuals from marrying.
XP is damn near 9 years old. That's close enough to 10.
Uh huh. But it wasn't three and a half years ago when Vista was released. Which is why statements about how "we just have this weird ten-year gap in operating systems" are wrong.
However, RAID has a huge failing in reliability. Every single time I've had a failure in a RAID setup, it was due to the controller -- your new single point of failure.
You're either incredibly lucky (to have never seen a disk failure) or incredibly unlucky (to have seen multiple controller failures so soon in your career).
But there's no standard (even for mirrored data) for the data on disk. So that means in a hardware RAID solution, you're rather screwed when the controller fails (unless you happened to have bought 2 -- firmware updates negate the possibility of purchasing in the future; I've run into that problem).
This is why you buy decent hardware, where the vendor will support you (and provide spares) for 3-5 years.
What I find interesting is that no one comments on this problem. The cheapo NAS devices typically use Linux, and probably even software RAID. I'd think they'd taught this in their marketing. But they tend to keep quiet about such things.
Controller failures are a vanishingly small proportion of events, especially when you're not buying cheap crap. In the last decade, I can only recall it happening 3 times, over thousands of systems I would have touched in that timeframe (and all of which had the vendor replace the controller with an identical one within a day). Disk failures - well, I see them a couple of times a _month_.
Regardless, the point of RAID is to keep data available. If a controller fails (unless you have multiple controllers and multipathing - pretty uncommon with local storage), then it's already failed at that - so whether or not you can get and identical replacement controller or not is a relatively minor issue, because you should be looking to your backups.
But really you have to properly label all the disks, preferably both in the on disk label and physically on the hard disk. That's kind of a pain, but it greatly reduces the likelihood of pulling the wrong disk.
I'm not aware of any remotely up-to-date software RAID system that requires the disks to be in the same order.
RAID is not peace of mind. Regular scheduled backups, with offsite storage, and tested recovery procedures is peace of mind. Its not considered backed up until your data been successfully restored from a backup.
Er, yes. Relevance to my comment being what, exactly ?
Slashdot Mirror
Microsoft designed a single user OS with no in built security in a time where networks were rare and have been forced to continue on with it by their customer base. All security ended up being tacked on because MS cant afford to kill legacy applications. I really don't think anyone at MS wants Windows to be insecure, it just happened that way and now they have to live with it.
Windows NT was designed from day one to be a multiuser OS. It's security is in no way "tacked on". The last version of Windows that fit your description was released a decade ago, and retired only a few years after that.
However, that's not the point. An AV is needed in Windows for the same reason it's needed on any other platform - it's a last ditch attempt at saving the end user when they've already circumvented the OS's security mechanisms. AV exists pretty much solely to protect the user from situations that OS security simply cannot.
That is simply absurd. If that were the case they would have few security flaws.
Do you have some numbers showing Windows has more flaws than other similar systems ?
A lot of commercial vendors treat independent researchers with contempt (how dare they find holes in our products) or as slaves (they should do the work our quality control dept should, for free)...
Of course, the folks who find a problem and then say "you have a week to fix this and then we release it into the wild" don't win their side any favours, either...
You didn't have to do this with Slackware in 1994.
Yes you did. Linux didn't get kernel modules until around 1995 (and they weren't in common usage until years after that).
Windows 7 doesn't even do this with hardware slightly older than it.
Windows 7 installs perfectly on my old Dell Precision M60 laptop, released in 2003. Not only that, but it actually works properly with my docking station and external monitors - something none of the popular Linux distributions manage to do.
some people have religious objections or objections to the organ donation system itself.
In which case they opt-out. Simple.
That's because the collateral damage from Microsoft's [...]
Thanks for not letting me down, Slashdot ! I knew someone would somehow manage to leverage this story for anti-Microsoft rhetoric.
This is probably a reference to this clause, found both in Vista and Windows 7 eulas. Available here in various eulas, such as Vista Home Premium English and Windows 7 Home Premium English, found here: microsoft Searching the XP sp2 eula does not seem to contain a similar clause.
So in other words, it bears only the vaguest and most indirect relevance to the OP's accusation. My, how surprising.
See their Trusted Platform Modules (TPM), DRM and HDCP divisions and reconsider our imaginary "freedom to run what we want" a double standard.
None of those do anything unless the application and/or user tells them to.
From Vista on a clause in every Windows EULA gives MS the right to delete executables and files from YOUR computer should they choose to do so (does XP also have this?)
Please quote the relevant section of the EULA.
Only upto 1996, and then Netscape moved to the free model because Microsoft was giving-away Internet Explorer.
Navigator wasn't made free until the beginning of 1998.
Bullshit. IE has never been superior to any other browser.
The industry disagrees. IE4 was rapidly stealing marketshare from Navigator long before it was integrated into Windows 98. Heck, the _beta_ version was winning non-trivial numbers of people away from Navigator 3 (and the freshly released 4).
As for the unholy disaster that was Navigator 4.0, it's hard to imagine anyone saying it was better than _anything_ with a straight face, let alone IE4. They'd nearly made it tolerable by 4.5 or so, but by then it was pretty much game over.
For that matter Windows has never been superior to any other OS. I can't think of a single product Microsoft has ever made where I would label it "the best" product on the market.
Millions of people clearly feel Windows is the best option for them. I certainly do, and it has nothing to do with lack of knowledge or experience with the alternatives.
They find one uncertainty or minor flaw in a study and suddenly volumes of studies -- even those unrelated -- can be thrown out and dismissed.
Tellingly, the same tactic is used by Creationists to try and discredit Evolution.
I will not tell it explicitly [...]
Because....?
Um, when Microsoft made that claim, they were referring to Windows XP and Internet Explorer 6 which are both almost 9 years old.
Actually it was about Windows 98 and IE4.
At that time, IE6 was very likely tightly linked to the OS.
It was no more "linked" then than it is now.
The argument is, and has always been, that the integration of IE into Windows resulted in various other parts of the system being dependent on the shared IE components to work. This statement is 100% true, and the subsequent emulation of that design by basically every other major platform has long since vindicated it from a technical perspective.
The only time the "IE is part of the OS" claim is incorrect, is when people deliberately choose a poor context in which to (mis)interpret it. IE is "part of the OS" the same way Explorer, Control Panel, the network stack, the Win32 API, or any number of other components are part of the OS. It was true then, it's true now, and it's also true of all the other major platforms as well (OSX w/Safari+Webkit, KDE w/Konquerer+khtml, GNOME w/Epiphany+Webkit).
But the big difference is OSX doesn't do anything to break FireFox or Opera.
Neither does Windows.
But a machine's administrator can easily replace /usr/bin/perl with any binary that implements a compatible interface.
As you could with IE.
Any /usr/bin/perl that parses and runs the same language will do, even if it has third-party defect fixes applied to it.
Unless, of course, there are programs that depend on some buggy behaviour a third-party replacement doesn't have.
Microsoft doesn't make that so easy with mshtml.dll: either you use Microsoft's mshtml.dll, or you don't use Windows.
Whether they make it easy or not is irrelevant (thought I fail to see how they make it "hard"). The principle - and therefore the argument - is the same.
Netscape was the best browser you could buy [...]
That's because it was pretty much the only browser you had to "buy".
IE was superior to Navigator in terms of function and performance from IE4 onwards, primarily because Navigator was an utter disaster from version 4.0 onwards. Even comparing IE 3.x and Navigator 3.x, the difference was so insignificant as to be purely a matter of preference or inertia.
Netscape lost because their product sucked. No other reason of consequence.
Each kid born, and each kid nursed, reduces the risk of breast cancer. It's a 5% drop and a 7% drop, or the other way around. Assuming every kid is nursed, that should be about 12% to 13% risk reduction per kid.
Pity about all those other life-threatening complications that can occur during pregnancy.
A woman naturally has about a dozen kids. Breast cancer is quite rare in countries where women birth early and often.
And what's the average life expectancy of women in those countries ?
I think the purpose homosexuality serves [...]
What is the basis of your assumption that it must serve a purpose ?
Homosexuality is likely a simple random mutation, which is why it shows up at roughly the same rate regardless of other factors, and why that rate doesn't change over time.
The value of marriage, compared to homosexuality, is that it usally leads to children, which will grow up and then pay taxes that are necessary sustain our standards of living in the long term.
Which is irrelevant, since homosexuals being married doesn't prevent any of that.
That's the basis for all pro-family / pro-marriage / etc. policies.
They're not "pro-marriage", they're "anti-gay". If they were just "pro-marriage", then they wouldn't give a damn about the gender of the two people getting married, since the likelihood of a married and unmarried gay couple producing children is the same, and has zero impact on whether or not a heterosexual couple will produce children (within marriage or otherwise).
It's not a moral debate, just a social one.
It's absolutely a moral debate. There's no social reason to prevent homosexuals from marrying.
I'll be amused if the kid, later in life, sues his parents for half the money.
With any luck, the parents would then sue back for all the money they spent raising him.
XP is damn near 9 years old. That's close enough to 10.
Uh huh. But it wasn't three and a half years ago when Vista was released. Which is why statements about how "we just have this weird ten-year gap in operating systems" are wrong.
However, RAID has a huge failing in reliability. Every single time I've had a failure in a RAID setup, it was due to the controller -- your new single point of failure.
You're either incredibly lucky (to have never seen a disk failure) or incredibly unlucky (to have seen multiple controller failures so soon in your career).
But there's no standard (even for mirrored data) for the data on disk. So that means in a hardware RAID solution, you're rather screwed when the controller fails (unless you happened to have bought 2 -- firmware updates negate the possibility of purchasing in the future; I've run into that problem).
This is why you buy decent hardware, where the vendor will support you (and provide spares) for 3-5 years.
What I find interesting is that no one comments on this problem. The cheapo NAS devices typically use Linux, and probably even software RAID. I'd think they'd taught this in their marketing. But they tend to keep quiet about such things.
Controller failures are a vanishingly small proportion of events, especially when you're not buying cheap crap. In the last decade, I can only recall it happening 3 times, over thousands of systems I would have touched in that timeframe (and all of which had the vendor replace the controller with an identical one within a day). Disk failures - well, I see them a couple of times a _month_.
Regardless, the point of RAID is to keep data available. If a controller fails (unless you have multiple controllers and multipathing - pretty uncommon with local storage), then it's already failed at that - so whether or not you can get and identical replacement controller or not is a relatively minor issue, because you should be looking to your backups.
Why are we still using a ball or laser on a cable to point on a screen where we want to do stuff?
Because no-one has come up with a better way.
RAID is for performance.
RAID is primarily about availability. The performance improvements are a nice side benefit.
To address your first point, RPOs and RTOs are _not_ the same thing as system availability. Backups address the former, RAID the latter.
But really you have to properly label all the disks, preferably both in the on disk label and physically on the hard disk. That's kind of a pain, but it greatly reduces the likelihood of pulling the wrong disk.
I'm not aware of any remotely up-to-date software RAID system that requires the disks to be in the same order.
RAID is not peace of mind. Regular scheduled backups, with offsite storage, and tested recovery procedures is peace of mind. Its not considered backed up until your data been successfully restored from a backup.
Er, yes. Relevance to my comment being what, exactly ?