The various Windows worms seem to be able to find an easy way onto any unpatched system remotely.
By _definition_ a worm can find it's way onto any unpatched system remotely. Maybe you need to compare them to linux "worms" instead of linux "vulnerabilities".
In this light, there's probably more driver support in most Linux distributions than 2003 Server provides in an officially supported manner.
So, which _server hardware_ are you trying to run Windows 2003 on that doesn't have drivers ?
I get the impression that people are complaining because Windows 2003 doesn't have "certified drivers" for their built-from-cheapest-parts $300 "server". To which my only response is: clearly decent hardware and reliability are not a major concern, so why are "certified" drivers ?
If you can't find Windows 2003 drivers for your system, "certified drivers" are probably the least of your problems.
Why would not being Administrator be a significant barrier for malware?
Strictly speaking it's not, as most malware doesn't really need higher privileges to do most of the things it does. However, a great deal of the stuff around today fails as soon as it can't write to parts of the registry and filesystem it expects to be able to.
You're missing my point. You said "its that in windows, almost anything can install a rootkit without antivirus or spyware proggies to stop it". This is not correct.
How then, is this any "easier" than linux or bsd?
OS X handles this situation best, although the simple fact is 90% of people will quite happily type in their password whenever prompted for it, making this sort of "protection" pretty weak.
Oh, and are you familiar with Shatter Attacks ?
Yes. It's a local privilege escalation attack that exploits applications that aren't properly written.
[...] but to this day, even a guest account on their graphical console can be used to subvert restrictions by the most trivial virus programs out there [...]
Funny how none of them do - I've yet to see a piece of malware that running as a non-Admin didn't stop.
What about hard core hackers who know the innards of windows and its sourcecode far more than my recently graduated roomie and his friends?
Then the situation is no different on Windows than it is on any other OS.
Anyways, a lot of what you say sounds like you get your training and education from the MCSE camp [...]
Not that it's relevant, but 90% of my "training and education" is from the coal face as a unix admin.
They will do what is easiest and most convenient.
This is precisely why 90% of "security problems" are the end user's fault and why _any_ platform with the prevalence of Windows would have similar issues.
You risk any competitive advantage the code gives you. This is a _huge_ issue to consider if the company is a market leader or experiencing strong growth due to the things its in-house software can do.
If some company is kicking arse and taking names in its market segment because of the advantages provided by in-house software, quite possibly one of the *dumbest* things that company could do would be to release that software to the world.
Also, IBM ended up wasting far too many resources on the OS/2 PPC port. Insiders have described it as one of the main reasons why OS/2 failed. Had the resources been put towards improving OS/2 and its hardware support, perhaps the majority of PC users today would be using OS/2 rather than XP or some other version of Windows.
This is somewhat believable when you consider OS/2 wasn't designed or written to be portable, but the biggest advantage Windows had over OS/2 was software availability - and that's because Microsoft were _very_ developer friendly and IBM were not. No amount of better hardware support or superior OS will help you if no-one is writing software for it.
Try getting drivers for Windows 2003 - that is MS's _only_ current supported server OS (2000 is out of support now). There is nowhere near the range of drivers as for XP. Some hardware you can find unofficial drivers or hacks, some just plain will not work. In fact, Linux probably has same or better driver support than Windows 2003.
Almost all Windows XP drivers work with Windows 2003 (as is to be expected, 2003 isn't a massive change from XP).
What server hardware can't you find Windows 2003 drivers for ?
Rootkits have always been a threat to admininstrators everywhere, but in windows, even spyware can autoinstall various programs, and it takes a lot of work and authorization from microshit to actually change anything in the OS without violating the EULA.
What are you trying to change - that you're supposed to be able to change - that you think you need special authorisation from Microsoft to do so ?
This may sound like a troll but think closely, and you'll see I'm right. It isn't the fact that an OS can have rootkits... its that in windows, almost anything can install a rootkit without antivirus or spyware proggies to stop it [...]
Simply not running as Administrator stops 99% of malware in its tracks (just like on any other multiuser OS).
This works best with microkernel architecture, which lets out Linux and Windows but OS X could conceivably go there. (And Windows actually could do it as well, since it is built around a kind of overblown microkernel.)
Unfortunately, OS X is about as much a "microkernel" as Windows NT is. Which is to say it was sort of a microkernel a long time ago, but practical requirements have long since moved most things into kernel space.
No, but I am part of one of the universities that got academic access to Windows source code. Although I didn't personally review it, my compsci department friends have horror stories.
Undoubtedly - most of them probably from the "Microsoft" part.
Windows is inherently insecure. At a fundamental level windows allows (and indeed, requires) too much access for programs to operate.
Evidence ?
Just look at how many applications need to be installed with administrator settings.
And then look at how many *properly written* programs achieve the same functionality _without_ requiring Administrator access.
This is an application developer problem, not a Windows problem.
To take an example, the only reason Doom 3 requires "Administrator access" because some idiot at id software decided to ignore the Windows standards for where configuration data should go, and store it in a.cfg file in the program directory. How is this sort of thing (it's extremely common) the fault of _Windows_ ? Would you blame Linux if some foolish developer decided that per-user configuration data would be stored in a file only writable by root ?
Compare to Unix (or Linux or OS X) and the clear separation of user and system.
Windows _has_ a clear separation of user and system. The problem is a large proportion of developers ignore it.
True. But that doesn't negative the fact that Windows is poorly designed.
And you saying "Windows is poorly designed" doesn't make it a fact, it makes it an opinion. If you want to make it a fact, say _why_ it's poorly designed and supply some actual supporting evidence for your arguments.
Friends and family using Windows==tons of trivial use questions. Friends and family using Mac==a few non trivial questions.
I'd be interested to see if that were actually true. I remember moving a relative from Windows to OS X - I had to answer exactly the same sort of "trivial questions" about OS X that I did about Windows.
Funny how if you use Mac OS X however you want, you don't have any security problems.
Really ? You think if I enabled the root account with a trivial password then fired up a few network services I wouldn't have any "security problems" ? How about if I ran exploit code as root, or simply typed in an admin password when prompted ? How about if I just ran a shell script that scanned through the system looking for email addresses and fired off some spam to any that it located ?
If/when OS X gets the same level of attention Windows does, it's going to have the same sorts of problems - because the vast bulk of "security problems" are caused by the end user doing silly things, not vulnerabilities in the OS.
If we could find an energy efficient way of moving water out of the earths gravity well (space elvator perhaps ?) then we could drench the martian surface.
Oh please, hitting your kids teaches them 2 things:
[...]
1) Actions have consequences.
2) Doing the wrong thing results in punishment.
From what I see, parents that hit their children only do it to releive their own stress and not because they really care, because if they did they would notice that their little Billy is a bully at school and constantly fights and their daughter looks for all the wrong male attention.
Funny how humans have managed to get this far without physical discipline destroying us. The whole "physical discipline == child abuse" meme is an/extremely/ recent development - and it doesn't appear to be providing much of a benefit to society as a whole.
Shatter is sort of interesting, but there are a few mitigating factors:
* It's largely an application developer issue
* It's not something that can be used for an automatic remote exploit
* Someone, at least, thinks they've found a complete "workaround".
* Since Win32 is being replaced by.NET, it's being addressed.
I'm not usually one for ontological arguments, but in this case....Windows was designed from the ground-up as a single user system.
This is not correct. NT was designed "from the ground up" to be multiuser. Indeed, with its pervasive ACLs, it's "more multiuser", if anything, than the typical unix (including OS X).
I would agree with you that OS X is not superior in "every aspect" but I wouldn't say that the rest is at most, a wash.
Do you really think you'll be seeing a standard ATX motherboard, power-supply and Pentium 4 in in the iMac form-factor and noise level?
No, but I never suggested you could.
I could certainly see a lightly modified micro-ATX or mini-ITX board fitting into an iMac form factor - not to mention a user-replaceable video card (the iMac's biggest weakness, IMHO).
Even if they change the looks of all the new Intel lineups, I highly doubt they're going to go with a form-factor that is nearly impossible to make look good (in my opinion, all PC cases are ugly).
There's no reason whatsover a PowerMac G5 couldn't be built in an ATX or BTX form-factor.
Machines like the iMac (or laptops), of course, require more customised boards - but the bulk of components remain off-the-shelf.
Seriously, maybe in their high end towers but their other product lines are going to require more than off the shelf components.
For all its form-factor coolness, the iMac is still made up of mostly bog-standard off-the-shelf components. The case and system board are the only custom-made components.
Heck, random people have come close to fitting mini-ITX form-factor boards into the Mac Mini - and that's with no access whatsoever to fabrication facilities or industrial designers.
My point is that there's nothing magical about Apple hardware. Certainly, some models are on the high end of the quality scale in terms of industrial design, but fundamentally Apple are doing the same thing Dell are and Macs don't have better hard disks, RAM, chipsets, etc, etc.
Apple kept throwing their own little 'next generation OS' parties, but nobody turned up.
No, _that_ isn't an even remotely accurate characterisation.
Microsoft _released_ their "next generation OS" in 1993.
Apple had a number of _aborted projects_, but nothing that made it to the release stage (until they bought NeXT and turned it into OS X).
Apple never had a party. They started to plan a few, maybe even sent out some invitations - but no party.
Or, to put it another way, "real artists ship";).
Of _course_ OS X has some technical superiorities over Windows - with 7 - 10 years more to watch and learn from the rest of the industry it would be nothing short of disgraceful if it didn't. However, the point so many people seem to ignore is that if Microsoft had started writing Windows NT in 1998 instead of 1988, it would probably look much like OS X does today. It's got nothing to do with better or worse development practices at Apple or Microsoft, or any inherent flaws, it's simply because Windows is a much older product with vastly more restrictions imposed by history than OS X is.
The wheel has turned. It'll turn again in the future. But there's a hell of a lot of people who either don't realise - or ignore - the fact that these things are cyclical.
A Mac is the entire machine. It starts with the case, and moves to the (usually Apple designed) motherboard, the Apple designed Bluetooth, the Apple designe Firewire, the Apple designed WiFi modules, then moves on to the placement of fans, the duct work, the attention to details. And you end up with a machine that performs the same function, but is of a much higher quality; a computer that's as much as computer as it is a piece of furnature in the room, and that's the idea of the Mac.
Fundamentally, Apple do nothing more than Dell (or any other whitebox supplier) does - they take a bunch of prefabricated parts from a number of sources and put them all together in their own way.
Like larger PC builders (eg: Dell, HP) a small number of their parts are in-house designed (eg: cases). The vast bulk of them, however, are off-the-shell third party components (RAM, CPUs, hard disks, video cards, etc).
Slashdot Mirror
By _definition_ a worm can find it's way onto any unpatched system remotely. Maybe you need to compare them to linux "worms" instead of linux "vulnerabilities".
So, which _server hardware_ are you trying to run Windows 2003 on that doesn't have drivers ?
I get the impression that people are complaining because Windows 2003 doesn't have "certified drivers" for their built-from-cheapest-parts $300 "server". To which my only response is: clearly decent hardware and reliability are not a major concern, so why are "certified" drivers ?
If you can't find Windows 2003 drivers for your system, "certified drivers" are probably the least of your problems.
Strictly speaking it's not, as most malware doesn't really need higher privileges to do most of the things it does. However, a great deal of the stuff around today fails as soon as it can't write to parts of the registry and filesystem it expects to be able to.
You're missing my point. You said "its that in windows, almost anything can install a rootkit without antivirus or spyware proggies to stop it". This is not correct.
How then, is this any "easier" than linux or bsd?
OS X handles this situation best, although the simple fact is 90% of people will quite happily type in their password whenever prompted for it, making this sort of "protection" pretty weak.
Oh, and are you familiar with Shatter Attacks ?
Yes. It's a local privilege escalation attack that exploits applications that aren't properly written.
[...] but to this day, even a guest account on their graphical console can be used to subvert restrictions by the most trivial virus programs out there [...]
Funny how none of them do - I've yet to see a piece of malware that running as a non-Admin didn't stop.
What about hard core hackers who know the innards of windows and its sourcecode far more than my recently graduated roomie and his friends?
Then the situation is no different on Windows than it is on any other OS.
Anyways, a lot of what you say sounds like you get your training and education from the MCSE camp [...]
Not that it's relevant, but 90% of my "training and education" is from the coal face as a unix admin.
They will do what is easiest and most convenient.
This is precisely why 90% of "security problems" are the end user's fault and why _any_ platform with the prevalence of Windows would have similar issues.
None of this things you've listed are even remotely guaranteed to occur. At best, they're optimistic assumptions.
If that doesn't work, you could point out that it won't cost them anything to GPL the code, but they stand to gain a lot.
Ot could cost them their competitive advantage in the market place. What sane businessman would give that up ?
The correct GNU/terminology is "freedom".
You risk any competitive advantage the code gives you. This is a _huge_ issue to consider if the company is a market leader or experiencing strong growth due to the things its in-house software can do.
If some company is kicking arse and taking names in its market segment because of the advantages provided by in-house software, quite possibly one of the *dumbest* things that company could do would be to release that software to the world.
This is somewhat believable when you consider OS/2 wasn't designed or written to be portable, but the biggest advantage Windows had over OS/2 was software availability - and that's because Microsoft were _very_ developer friendly and IBM were not. No amount of better hardware support or superior OS will help you if no-one is writing software for it.
Almost all Windows XP drivers work with Windows 2003 (as is to be expected, 2003 isn't a massive change from XP).
What server hardware can't you find Windows 2003 drivers for ?
What do you think an animal is ?
Or, as they said on Dr Who, "life is nature's way of keeping meat fresh".
I want to know why it's ethical to kill plants, but not ethical to kill animals.
What are you trying to change - that you're supposed to be able to change - that you think you need special authorisation from Microsoft to do so ?
This may sound like a troll but think closely, and you'll see I'm right. It isn't the fact that an OS can have rootkits... its that in windows, almost anything can install a rootkit without antivirus or spyware proggies to stop it [...]
Simply not running as Administrator stops 99% of malware in its tracks (just like on any other multiuser OS).
Unfortunately, OS X is about as much a "microkernel" as Windows NT is. Which is to say it was sort of a microkernel a long time ago, but practical requirements have long since moved most things into kernel space.
Even without Office, Outlook is "essentially free" - every Exchange CAL includes an Outlook license.
Undoubtedly - most of them probably from the "Microsoft" part.
Windows is inherently insecure. At a fundamental level windows allows (and indeed, requires) too much access for programs to operate.
Evidence ?
Just look at how many applications need to be installed with administrator settings.
And then look at how many *properly written* programs achieve the same functionality _without_ requiring Administrator access.
This is an application developer problem, not a Windows problem.
To take an example, the only reason Doom 3 requires "Administrator access" because some idiot at id software decided to ignore the Windows standards for where configuration data should go, and store it in a .cfg file in the program directory. How is this sort of thing (it's extremely common) the fault of _Windows_ ? Would you blame Linux if some foolish developer decided that per-user configuration data would be stored in a file only writable by root ?
Compare to Unix (or Linux or OS X) and the clear separation of user and system.
Windows _has_ a clear separation of user and system. The problem is a large proportion of developers ignore it.
True. But that doesn't negative the fact that Windows is poorly designed.
And you saying "Windows is poorly designed" doesn't make it a fact, it makes it an opinion. If you want to make it a fact, say _why_ it's poorly designed and supply some actual supporting evidence for your arguments.
Friends and family using Windows==tons of trivial use questions. Friends and family using Mac==a few non trivial questions.
I'd be interested to see if that were actually true. I remember moving a relative from Windows to OS X - I had to answer exactly the same sort of "trivial questions" about OS X that I did about Windows.
Funny how if you use Mac OS X however you want, you don't have any security problems.
Really ? You think if I enabled the root account with a trivial password then fired up a few network services I wouldn't have any "security problems" ? How about if I ran exploit code as root, or simply typed in an admin password when prompted ? How about if I just ran a shell script that scanned through the system looking for email addresses and fired off some spam to any that it located ?
If/when OS X gets the same level of attention Windows does, it's going to have the same sorts of problems - because the vast bulk of "security problems" are caused by the end user doing silly things, not vulnerabilities in the OS.
Go to Mars and burn some hydrogen ?
That would be the people writing them.
Just like Red Hat, Sun, Apple and everyone else who sells software, you mean ?
[...]
1) Actions have consequences.
2) Doing the wrong thing results in punishment.
From what I see, parents that hit their children only do it to releive their own stress and not because they really care, because if they did they would notice that their little Billy is a bully at school and constantly fights and their daughter looks for all the wrong male attention.
Funny how humans have managed to get this far without physical discipline destroying us. The whole "physical discipline == child abuse" meme is an /extremely/ recent development - and it doesn't appear to be providing much of a benefit to society as a whole.
Shatter is sort of interesting, but there are a few mitigating factors:
* It's largely an application developer issue
* It's not something that can be used for an automatic remote exploit
* Someone, at least, thinks they've found a complete "workaround".
* Since Win32 is being replaced by .NET, it's being addressed.
I'm not usually one for ontological arguments, but in this case....Windows was designed from the ground-up as a single user system.
This is not correct. NT was designed "from the ground up" to be multiuser. Indeed, with its pervasive ACLs, it's "more multiuser", if anything, than the typical unix (including OS X).
I would agree with you that OS X is not superior in "every aspect" but I wouldn't say that the rest is at most, a wash.
Why not ?
No, but I never suggested you could.
I could certainly see a lightly modified micro-ATX or mini-ITX board fitting into an iMac form factor - not to mention a user-replaceable video card (the iMac's biggest weakness, IMHO).
Even if they change the looks of all the new Intel lineups, I highly doubt they're going to go with a form-factor that is nearly impossible to make look good (in my opinion, all PC cases are ugly).
There's no reason whatsover a PowerMac G5 couldn't be built in an ATX or BTX form-factor.
Machines like the iMac (or laptops), of course, require more customised boards - but the bulk of components remain off-the-shelf.
Seriously, maybe in their high end towers but their other product lines are going to require more than off the shelf components.
For all its form-factor coolness, the iMac is still made up of mostly bog-standard off-the-shelf components. The case and system board are the only custom-made components.
Heck, random people have come close to fitting mini-ITX form-factor boards into the Mac Mini - and that's with no access whatsoever to fabrication facilities or industrial designers.
My point is that there's nothing magical about Apple hardware. Certainly, some models are on the high end of the quality scale in terms of industrial design, but fundamentally Apple are doing the same thing Dell are and Macs don't have better hard disks, RAM, chipsets, etc, etc.
No, _that_ isn't an even remotely accurate characterisation.
Microsoft _released_ their "next generation OS" in 1993.
Apple had a number of _aborted projects_, but nothing that made it to the release stage (until they bought NeXT and turned it into OS X).
Apple never had a party. They started to plan a few, maybe even sent out some invitations - but no party.
Or, to put it another way, "real artists ship" ;).
Of _course_ OS X has some technical superiorities over Windows - with 7 - 10 years more to watch and learn from the rest of the industry it would be nothing short of disgraceful if it didn't. However, the point so many people seem to ignore is that if Microsoft had started writing Windows NT in 1998 instead of 1988, it would probably look much like OS X does today. It's got nothing to do with better or worse development practices at Apple or Microsoft, or any inherent flaws, it's simply because Windows is a much older product with vastly more restrictions imposed by history than OS X is.
The wheel has turned. It'll turn again in the future. But there's a hell of a lot of people who either don't realise - or ignore - the fact that these things are cyclical.
Hardly. NT and VMS share architects and - to a certain degree - development teams. OS X is a development of the NeXT _codebase_.
OS X is to NeXT as, say, Windows 2003 is to Windows NT 3.1.
Which Pentium M are you thinking of that's faster than a dual core P4 ? Or even the fastest P4, for that matter ?
Fundamentally, Apple do nothing more than Dell (or any other whitebox supplier) does - they take a bunch of prefabricated parts from a number of sources and put them all together in their own way.
Like larger PC builders (eg: Dell, HP) a small number of their parts are in-house designed (eg: cases). The vast bulk of them, however, are off-the-shell third party components (RAM, CPUs, hard disks, video cards, etc).
Apple == Dell+Microsoft. Get over it.