Microsoft products look like their built by a giant, complacent bureaucracy where the right hand doesn't know what the left is doing.
(now why would that be?)
They should have stuck with making the best compilers in the business, or at least stuck to their goal of upgrading DOS until it was functionally capable of replacing their high end Xenix product... when they started selling vaporware (they were advertising Windows in 1983, even before the Mac was out, and didn't really have a usable product until Moore's Law saved them in the '90s) they switched from being a company that's all about making great software into one that's all about selling great-looking software.
I said this years ago, when Microsoft dodged the break-up bullet: they'd be better off in the long run being broken up, than trying to run a business that was all things to all people
I think arevos has pretty much covered my reasons for preferring Javascript's object model. Complexity is a huge huge huge red flag for me. I also dislike bondage-and-discipline type structures, and prefer dynamic typing, but that's a whole other can of toxic waste I don't want to open.
I think the biggest problem in QM is the idea that the "collapse of the state vector" actually describes anything real. It's one of those questions like "when does life start" or "what's really a planet" that doesn't really have anything to do with science. It's just a metaphor that makes certain kinds of reasoning about QM easier, and provides guidance as to where you can simplify your model to make the calculations practical.
ECMAScript 2 (Javascript) has a better object model than Java. Unfortunately ECMAScript 3 is giving up a lot of the advantages in an attempt to squash problems that are at least as much the fault of embedded libraries as the object model.:(
I'd really like to see FScript opened up to more than just the Mac. That's the thing that really makes me reluctant to get into it on the Mac... I don't want to work in a language (open source or not) that isn't portable.
But raytracing doesn't do realistic lighting at all!
It does more realistic lighting than rasterization, and it definitely will do caustics... you just need to shoot more rays. Whether you can shoot enough rays in realtime or not, well, that's where you need the speedup from an RPU.
"Global Illumination" isn't a lighting effect, it's a heuristic for rasterizing that fakes some effects that require additional rays to calculate. In some cases that's ludicrously many rays, in others it's not. There's also some very good (albeit still expensive) techniques to simulate radiance and other "global illumination" effects in raytracing. Again, how much you can do in realtime depends on your resources. And, again, an RPU would be a major win.
Large diagrams, tables, complex multi-page descriptions, and so on would be where you would pick up the biggest advantage from this. If you're referring to multiple pages you'd simply leave them separated.
Yes, a compromised Mail.app can send spam. But that gains you nothing if you compromised Safari.
Mail.app is also one of the applications being sandboxed... and a compromised Mail.app can do pretty much anything a compromised Safari can do, since Mail.app is able to display embedded images.
In the real world, there are still a number of obstacles to take, including protections by the webmail providers and network delays that are likely to reduce your spam-rate beneath the amount useful.
Like canceling your webmail account.
Are you for real? There's a reason that banks use TAN lists
How many banks actually use them? I suspect it must be fewer than you think.
One, what's to say against saving only into the download directory? You can copy files from there. It's a tiny bit less convenient, but not much.
That's one, and it's acknowledging my point. What's two?
I don't know about many plugins that require write privileges to your file system.
A lot of plugins are just alternate "hooks" into desktop applications.
I've given people root shells on SELinux machines and told them to try and break stuff.
You don't have to "break stuff" to take advantage of a compromised application. I've listed a number of attacks that a compromised Firefox and Mail.app can perform that you have agreed are credible (yes, you did agree that half the attacks I was describing were possible) that don't even require you to be able to "break stuff".
A sandbox - or almost any other security thing - is just bandaid we plaster around a broken software because we can't fix it.
And it gives the creator of the broken software an excuse not to fix it.
Even if the firewall Microsoft is creating around IE wasn't leaky, it wouldn't prevent a compromised IE from being abused within the restrictions of the firewall, and it IS being used by Microsoft as an excuse NOT to fix the real problems.
Realistic lighting allows you to use those clever algorithms in your head that you've learned over the past 20+ years in the real world, so when you see a flicker of a reflection or a change in the shadows in a darkened tunnel you can turn and blast the damn camper on the opposite rooftop before he nails you with his sniper rifle.
Seems like the logical thing to do would be to simply allow multiple readers to cooperate in document display, so when you snap together (say) 4 readers, you get a 4-page view, split them apart you get 2 2-page views of different documents. You would use proximity sensors to define which pages were "together", and simple mechanical clips to hold them together when not laying flat. You could then hand one side of the page you were looking at to a co-worker, then pull a blank sheet out of your drawer to restore your own reading area, while he walks off with the other page.
This is becoming more like an extension of the Xerox PAD and TAB.
The documents mightn't be stored on the device, rather they would be accessed via the office wifi network.
Now the issue here is that if the certificate has been modified during transit, then your security has been breached. This is something solved with CA signed certificates.
It's also solved by having the browser install the unsigned certificate, which would allow it to go "OK, you're accessing this website through the same certificate as last time, nothing to see here", or to go "DANGER DANGER, the unsigned certificate for this site has changed!"
This is what SSH does with keyfiles, why don't browsers do this with certificates?
Because the certificate authority says they're not supposed to?
Or because nobody's thought of this obvious improvement?
Do you know how ABS works? If you did, you'd know they are anything but smooth.
I've used them. I'm not just a commentator, I'm one of the automotive lusers I'm talking about: I found myself braking shorter and depending on ABS to bring me to a smooth stop (yes, you can feel it when it kicks in, that's not what I mean by "smooth") if I happened to go outside the envelope. Then I read the report, realized they were talking about me and modified my behavior.
Any security feature, be it ABS or sandboxes, must be judged in the context of how its presence modified human behavior.
Current lamer (not even cracker) scene on OS X was therotically speaking about modification of binaries to inject their code, exploiting input manager functionality to attach to running code and do some tricks with their keychains (of course). Apple as OS vendor proactively did some stuff to prevent such lame but dangerous stuff.
Ah yes, that's another piece of security theatre. You can still perform Mach injection on applications under Leopard, the technique is just changed slightly.
There is no way someone can tamper a executable on my OS but on average Mac users machine (which they target), they will try doing it.
There are many useful tools that non-technical users use that were broken by Leopard, and everything that Apple was trying to prevent by breaking those tools has been worked around.
So, anyway, I assume that you're referring to people like rosyna as "lamers". I strongly disagree, but I'll keep your bias in mind.
The code injection prevention, watch of OS folders, watching for suspicious activity is a job of true heuristic antivirus.
Antivirus software is also security theatre. I don't use it, even on Windows, and the last time I had to remove a virus on any computer I own was in the early '90s.
It's the difference between "the application Mail has been updated. Do you want it to allow access to your keychain items?" and not bugging you at all.
According to the quoted site, it will do that every time I run the app.
I would rather have it ask me when I update the application, which is a rare event, than have it silently trust a modified application because it has been signed with a key it trusts.
Safari can't send mail. If you click on a mailto: link in Safari, it will launch Mail.app.
You're picking nits here.
* If Mail.app can send mail, a compromised Mail.app can send spam.
* If Safari can be used to send mail through your webmail servers, it can send spam through them.
* If Safari can open my bank's web page, it can transfer money out of my account.
Nonsense for any real-world bank. With absolutely no exception, actual banks use additional security, such as TAN lists.
Picking nits again.
* If you can make payments through your bank's web page, Safari can transfer money out of your account.
In a sandboxed environment, you would have a specific download directory (you already do, but bear with me) and Safari would have write priviledges to that directory only.
So "Save File as" is one of those "conveniences" you have to give up.
You also have to give up using many plugins.
that's a far cry from the "total destruction" scenario you outline.
Straw man: I'm not outlining a "total destruction" scenario, I'm pointing out that a sandbox does not protect you from a compromised application.
[A sandbox] is much, much better than doing nothing.
There are other alternatives than deploying a sandbox or doing nothing, and a sandbox that leaves so many attacks open is something I would rather forgo because it will encourage application writers to depend on the sandbox rather than making the kinds of changes that REALLY need to be made. And that's where we need to concentrate on: fixing the applications to remove the really dangerous features... not things like "save as", but things like "Open 'Safe' files after downloading". I hope that if you use Safari you've checked that one's been turned off.
The real "Next Step for Mac (& Windows) Users"
on
Two Trojans For Mac OS X
·
· Score: 4, Informative
History shows us that even the smartest of users can catch malware.
It's been 17 years since the last time I had to remove a virus from my own computer, even when that computer's been unpatched Windows 2000 connected to the Internet. In the years that I was network and security admin and had control of the network, the only time we had any systems infected was when a user had either downloaded and run a file (that is, they were social-engineered, and in 10 years only one person came to me with an infected laptop after doing that twice) or they had violated my policy banning IE and Outlook at our location.
The potential for infection if you avoid software that supports automatic execution of remote content is very very small, even on Windows. The reason that Windows has a high infection rate is because of IE and Outlook, not simply because it's popular.
If you're on a Mac, and use Safari, here's the next steps you should take:
(1) Go into preferences and make sure "Open 'Safe' Files after Downloading" is disabled. (2) Get a standalone FTP client and use one of the third-party LaunchServices editors (look for internet access preference panes) and change the default application for FTP: URLs from Finder to something else. (3) Use Tinkertool or equivalent to disable Dashboard.
#1 is the most important. #2 and #3 don't allow automatic execution of untrusted content, but they do make social engineer ing easier.
If you use a Gecko-based browser like Firefox or Camino, you don't need to worry about these.
If you're on Windows: avoid using any application that uses the Microsoft HTML control to access untrusted content. That includes IE, Outlook (not all versions, any more, but I believe you have to accept the Vista-style UI to avoid it), Windows Media Player, Realplayer, and some Firefox plugins and some versions of Netscape.
In Firefox, Windows or Mac or Linux, always clean out the whitelist for installing extensions after you install an extension... the installer is an autoexecution mechanism, and there have been exploits that took advantage of that even if you don't approve the install dialog.
The scary part is that most Mac OS users think they can't catch malware because they're smart enough not to install it.
At the moment that's not far from the truth. You can avoid catching malware by being smart enough to avoid running it, on Windows or OS X, if you exercise some care in the applications you use, and how they're configured. It's harder on Windows, but it's still possible.
These trojans are purely payload. The delivery mechanism is still social engineering... not remote execution. We know that "once you're penetrated you're ****ed", pointing out again the ways you can be ****ed is not news (for nerds or otherwise) nor stuff that matters.
These are not the viruses you're looking for. Nothing to see here, move along.
I'm so damn sick of people going "oooh, aaah, I thought $software was immune to $threat" when no credible commentator has made such a claim.
Just quit it, OK? It just makes you look like an utter twit.
And it's not just a lack of being targeted. It's a smaller surface area for attack, as well. OS X has nothing comparable to the rich viral petrie dish that the tight desktop-browser integration in Windows provides. Before 1997, Windows viruses were virtually all a matter of tricking people into running software, not having software automatically run when you just select an email message so you can delete it... which is how bad things were in the late '90s. Microsoft has tightened up the gaping holes in Windows since then, but they have done NOTHING to remove the underlying flaw that makes these kinds of attacks so easy there.
Compared to Windows, OS X is "virus resistant". That doesn't mean "virus proof". But it does mean that it's going to remain harder to infect than Windows until such time as Apple decides to implement something as barking mad as ActiveX.
Slashdot Mirror
This is obviously not belonging to "Your Rights Online".
The second amendment obviously covers online munitions as well, which are known to include cryptography and intrusion detection systems.
Microsoft products look like their built by a giant, complacent bureaucracy where the right hand doesn't know what the left is doing.
(now why would that be?)
They should have stuck with making the best compilers in the business, or at least stuck to their goal of upgrading DOS until it was functionally capable of replacing their high end Xenix product... when they started selling vaporware (they were advertising Windows in 1983, even before the Mac was out, and didn't really have a usable product until Moore's Law saved them in the '90s) they switched from being a company that's all about making great software into one that's all about selling great-looking software.
I said this years ago, when Microsoft dodged the break-up bullet: they'd be better off in the long run being broken up, than trying to run a business that was all things to all people
<AOL>Me too</AOL>
I think arevos has pretty much covered my reasons for preferring Javascript's object model. Complexity is a huge huge huge red flag for me. I also dislike bondage-and-discipline type structures, and prefer dynamic typing, but that's a whole other can of toxic waste I don't want to open.
I think the biggest problem in QM is the idea that the "collapse of the state vector" actually describes anything real. It's one of those questions like "when does life start" or "what's really a planet" that doesn't really have anything to do with science. It's just a metaphor that makes certain kinds of reasoning about QM easier, and provides guidance as to where you can simplify your model to make the calculations practical.
ECMAScript 2 (Javascript) has a better object model than Java. Unfortunately ECMAScript 3 is giving up a lot of the advantages in an attempt to squash problems that are at least as much the fault of embedded libraries as the object model. :(
I'd really like to see FScript opened up to more than just the Mac. That's the thing that really makes me reluctant to get into it on the Mac... I don't want to work in a language (open source or not) that isn't portable.
But raytracing doesn't do realistic lighting at all!
It does more realistic lighting than rasterization, and it definitely will do caustics... you just need to shoot more rays. Whether you can shoot enough rays in realtime or not, well, that's where you need the speedup from an RPU.
"Global Illumination" isn't a lighting effect, it's a heuristic for rasterizing that fakes some effects that require additional rays to calculate. In some cases that's ludicrously many rays, in others it's not. There's also some very good (albeit still expensive) techniques to simulate radiance and other "global illumination" effects in raytracing. Again, how much you can do in realtime depends on your resources. And, again, an RPU would be a major win.
Large diagrams, tables, complex multi-page descriptions, and so on would be where you would pick up the biggest advantage from this. If you're referring to multiple pages you'd simply leave them separated.
is there a reason my reader shouldn't be able to display the same pages at the same time as yours?
Of course not. I'm not sure why you'd think that would be an issue, I certainly don't think I suggested it.
I've been asking this since I first heard about the SSL certificate model, let alone saw it in action.
I've yet to see a good explanation.
Yes, a compromised Mail.app can send spam. But that gains you nothing if you compromised Safari.
Mail.app is also one of the applications being sandboxed... and a compromised Mail.app can do pretty much anything a compromised Safari can do, since Mail.app is able to display embedded images.
In the real world, there are still a number of obstacles to take, including protections by the webmail providers and network delays that are likely to reduce your spam-rate beneath the amount useful.
Like canceling your webmail account.
Are you for real? There's a reason that banks use TAN lists
How many banks actually use them? I suspect it must be fewer than you think.
One, what's to say against saving only into the download directory? You can copy files from there. It's a tiny bit less convenient, but not much.
That's one, and it's acknowledging my point. What's two?
I don't know about many plugins that require write privileges to your file system.
A lot of plugins are just alternate "hooks" into desktop applications.
I've given people root shells on SELinux machines and told them to try and break stuff.
You don't have to "break stuff" to take advantage of a compromised application. I've listed a number of attacks that a compromised Firefox and Mail.app can perform that you have agreed are credible (yes, you did agree that half the attacks I was describing were possible) that don't even require you to be able to "break stuff".
A sandbox - or almost any other security thing - is just bandaid we plaster around a broken software because we can't fix it.
And it gives the creator of the broken software an excuse not to fix it.
Even if the firewall Microsoft is creating around IE wasn't leaky, it wouldn't prevent a compromised IE from being abused within the restrictions of the firewall, and it IS being used by Microsoft as an excuse NOT to fix the real problems.
You wouldn't put them together like that. I'm not talking about books, I'm talking about workspaces.
Talk about your latitude adjustment.
Well you are, OK, that was funny.
But it's also serious.
GOD DAMN the Word document structure sucks like something that sucks a lot.
Instead of a video card you would just need a faster cpu, which if we base off of moore's law won't be much longer.
If the video card makers had picked up on the RPU you could use your video card to get realistic high frame-rate raytraced games today.
Dr Slusallek is working at nVidia now, so who knows?
Realistic lighting allows you to use those clever algorithms in your head that you've learned over the past 20+ years in the real world, so when you see a flicker of a reflection or a change in the shadows in a darkened tunnel you can turn and blast the damn camper on the opposite rooftop before he nails you with his sniper rifle.
Seems like the logical thing to do would be to simply allow multiple readers to cooperate in document display, so when you snap together (say) 4 readers, you get a 4-page view, split them apart you get 2 2-page views of different documents. You would use proximity sensors to define which pages were "together", and simple mechanical clips to hold them together when not laying flat. You could then hand one side of the page you were looking at to a co-worker, then pull a blank sheet out of your drawer to restore your own reading area, while he walks off with the other page.
This is becoming more like an extension of the Xerox PAD and TAB.
The documents mightn't be stored on the device, rather they would be accessed via the office wifi network.
If you can use your Macbook Pro on your lap you've got tougher *legs* than I do, let alone anything you keep between them.
Now the issue here is that if the certificate has been modified during transit, then your security has been breached. This is something solved with CA signed certificates.
It's also solved by having the browser install the unsigned certificate, which would allow it to go "OK, you're accessing this website through the same certificate as last time, nothing to see here", or to go "DANGER DANGER, the unsigned certificate for this site has changed!"
This is what SSH does with keyfiles, why don't browsers do this with certificates?
Because the certificate authority says they're not supposed to?
Or because nobody's thought of this obvious improvement?
Do you know how ABS works? If you did, you'd know they are anything but smooth.
I've used them. I'm not just a commentator, I'm one of the automotive lusers I'm talking about: I found myself braking shorter and depending on ABS to bring me to a smooth stop (yes, you can feel it when it kicks in, that's not what I mean by "smooth") if I happened to go outside the envelope. Then I read the report, realized they were talking about me and modified my behavior.
Any security feature, be it ABS or sandboxes, must be judged in the context of how its presence modified human behavior.
Current lamer (not even cracker) scene on OS X was therotically speaking about modification of binaries to inject their code, exploiting input manager functionality to attach to running code and do some tricks with their keychains (of course). Apple as OS vendor proactively did some stuff to prevent such lame but dangerous stuff.
Ah yes, that's another piece of security theatre. You can still perform Mach injection on applications under Leopard, the technique is just changed slightly.
There is no way someone can tamper a executable on my OS but on average Mac users machine (which they target), they will try doing it.
There are many useful tools that non-technical users use that were broken by Leopard, and everything that Apple was trying to prevent by breaking those tools has been worked around.
So, anyway, I assume that you're referring to people like rosyna as "lamers". I strongly disagree, but I'll keep your bias in mind.
The code injection prevention, watch of OS folders, watching for suspicious activity is a job of true heuristic antivirus.
Antivirus software is also security theatre. I don't use it, even on Windows, and the last time I had to remove a virus on any computer I own was in the early '90s.
It's the difference between "the application Mail has been updated. Do you want it to allow access to your keychain items?" and not bugging you at all.
According to the quoted site, it will do that every time I run the app.
I would rather have it ask me when I update the application, which is a rare event, than have it silently trust a modified application because it has been signed with a key it trusts.
History shows us that even the smartest of users can catch malware.
It's been 17 years since the last time I had to remove a virus from my own computer, even when that computer's been unpatched Windows 2000 connected to the Internet. In the years that I was network and security admin and had control of the network, the only time we had any systems infected was when a user had either downloaded and run a file (that is, they were social-engineered, and in 10 years only one person came to me with an infected laptop after doing that twice) or they had violated my policy banning IE and Outlook at our location.
The potential for infection if you avoid software that supports automatic execution of remote content is very very small, even on Windows. The reason that Windows has a high infection rate is because of IE and Outlook, not simply because it's popular.
If you're on a Mac, and use Safari, here's the next steps you should take:
(1) Go into preferences and make sure "Open 'Safe' Files after Downloading" is disabled.
(2) Get a standalone FTP client and use one of the third-party LaunchServices editors (look for internet access preference panes) and change the default application for FTP: URLs from Finder to something else.
(3) Use Tinkertool or equivalent to disable Dashboard.
#1 is the most important. #2 and #3 don't allow automatic execution of untrusted content, but they do make social engineer ing easier.
If you use a Gecko-based browser like Firefox or Camino, you don't need to worry about these.
If you're on Windows: avoid using any application that uses the Microsoft HTML control to access untrusted content. That includes IE, Outlook (not all versions, any more, but I believe you have to accept the Vista-style UI to avoid it), Windows Media Player, Realplayer, and some Firefox plugins and some versions of Netscape.
In Firefox, Windows or Mac or Linux, always clean out the whitelist for installing extensions after you install an extension... the installer is an autoexecution mechanism, and there have been exploits that took advantage of that even if you don't approve the install dialog.
The scary part is that most Mac OS users think they can't catch malware because they're smart enough not to install it.
At the moment that's not far from the truth. You can avoid catching malware by being smart enough to avoid running it, on Windows or OS X, if you exercise some care in the applications you use, and how they're configured. It's harder on Windows, but it's still possible.
These trojans are purely payload. The delivery mechanism is still social engineering... not remote execution. We know that "once you're penetrated you're ****ed", pointing out again the ways you can be ****ed is not news (for nerds or otherwise) nor stuff that matters.
These are not the viruses you're looking for. Nothing to see here, move along.
I'm so damn sick of people going "oooh, aaah, I thought $software was immune to $threat" when no credible commentator has made such a claim.
Just quit it, OK? It just makes you look like an utter twit.
And it's not just a lack of being targeted. It's a smaller surface area for attack, as well. OS X has nothing comparable to the rich viral petrie dish that the tight desktop-browser integration in Windows provides. Before 1997, Windows viruses were virtually all a matter of tricking people into running software, not having software automatically run when you just select an email message so you can delete it... which is how bad things were in the late '90s. Microsoft has tightened up the gaping holes in Windows since then, but they have done NOTHING to remove the underlying flaw that makes these kinds of attacks so easy there.
Compared to Windows, OS X is "virus resistant". That doesn't mean "virus proof". But it does mean that it's going to remain harder to infect than Windows until such time as Apple decides to implement something as barking mad as ActiveX.