I haven't even customized my user agent string and I'm using the standard Fedora 14 browser, but my user agent string itself is unique... Seems like I am the only Danish Fedora 14 user who has clicked on panopticlick recently.
One wonders what your position is regarding science and global warming.
Easy: the evidence indicates that our CO2 emissions are causing climate change. Caution should therefore lead us to reduce emissions until such time that we can be reasonably sure that we were wrong and CO2 emissions are safe.
Once we are reasonably sure that we can engineer GM crops to not spread their modified genes to other plants or become invasive species, we can reverse the ban. Assuming they do not compromise the security of our food sources in other ways -- e.g. monoculture is a risk whether it is GM or not.
Yes, neutrinos are more common near nuke plants. At least that is what theory tells us. If you find a cheap way to PROVE this experimentally, you would become moderately famous among physicists. Getting extra glitches from memory would qualify...
You cannot encrypt a password both in transfer and on disk (unless you use a separately encrypted channel with separate authentication, but then why do you need a password?). For a lot of things it is more important that you can use the password to establish a secure channel than it is to store the password as a hash.
E.g. with the simple "ask for three specific characters from the password" method you gain almost-one-time-passwords, so a keylogger on a public terminal cannot empty your bank account afterwards. This property is probably worth the risk that someone breaks into the password storage.
So they're saying that enabling shader code execution allows web sites to exploit hypothetical vulnerabilities in the graphics driver?
They're not particularly hypothetical. Graphics driver code is such that games programmers carefully work around bugs in order to not crash anything. Imagine if every program running on the main CPU had to carefully avoid certain instruction sequences in order to not crash the system -- would you run a multi-user system on that?
Then again, that was how it was in the 80's on many time sharing systems...
It is probably not the cable provider doing it, but the content providers. They like to make demands that you only offer a particular channel to subscribers in a package with these other channels, or that you can only get this channel if you provide it to ALL your subscribers.
The currently standing record for GSM "decryption" is to pretend to be the provider and tell the phone to not encrypt. Sure, that does not involve any actual decryption, but since you as a user cannot tell it happened, what exactly is the difference?
You people keep repeating the same mistake over and over!
There is no problem sharing IP 2.2.2.2 port 34567 between thousands of connections. It is only a problem if all those users decide to connect to the same server on the same port simultaneously. This is highly unlikely, and few servers could handle that anyway. Google is NOT a counterexample, they use lots of IP addresses at their end.
you do realize that that the port range is a 16-bit integer?
You do realize that proper NAT uses the whole 5-tuple for flow identification? The 16-bit port range is only a problem for old Cisco implementations of NAT; I am not aware of any non-Cisco devices with such a braindead design.
Nokia had smart phones basically since 9600bps became possible over GSM. They were heavy and awkward to use, of course, but the Windows CE phones were worse, so that is hardly an argument in favour of Microsoft.
The Icelandic government didn't borrow anything. The private bank Icesave borrowed a lot of money from banks and individuals primarily in the UK and the Netherlands. It went bankrupt, and the Dutch and UK governments decided to pay the banks and individuals what Icesave owed them. Very kind of those governments. Afterwards they decided they wanted the money from the Icelandic government.
Is it any surprise that the Icelandic public is failing to see why they should bail out private investors and banks in other countries, when no guarantee was issued in the first place? I am sure the US public is having similar difficulties with why they had to bail out the US credit institutions, but at least in that case they could vote out the politicians who did it.
An application could place its own window right at the place where I am expecting security-sensitive information to appear. Like right above the security lock icon in Firefox, making me believe that a particular site has a valid certificate. It is even worse if that affects focus, obviously.
And yes, X and the current window managers are broken in precisely this way right now. It does not matter too much because everyone implicitly trusts their X clients. In phone OS's, there is no such implicit trust, and we need to get there on the desktop too.
If you made your mirrors out of Faraday isolators, couldn't you theoretically have a mirror which admitted light from behind while reflecting light from the front? That ought to enable you to beat the concentration limit.
Why not let your window manager do the job it was designed to do?
Because none of the common window managers actually do a decent job of managing windows. In turn, no one uses overlapping windows, instead switching between full-screen applications.
What really annoys me is that it is so trivially solved: the system should not raise (or lower or otherwise arrange, or map or unmap) a window EVER. The application must do it in response to events
That is completely the wrong way around. The application should stick to its own business and run completely sandboxed in its own window set, and let the user handle the placement and order. It is bloody annoying when Firefox manages to hijack the whole desktop and you cannot even switch virtual desktop. Applications should NOT be able to make modal dialog boxes. Allowing applications that kind of control is a security risk which is only acceptable because we have almost no sandboxing at all in all major desktop operating systems.
Phone OS's seem to be doing a lot better on the security side, and once the same kind of security is applied to the desktop, the windowing model will need to be secured. Phone OS's tend to handle the windowing model by not allowing overlapping windows at all.
You have never had to re-type something from one window to another, or simply had to react to something going on in one window by typing in another window? The first would be fixed by working copy-paste, but alas that seems to be getting increasingly broken even in Linux. Back in Red Hat 3 you could copy-paste all text except window titles; now some dialog boxes have text which cannot be copied.
I was totally happy with sawfish on 7.3, and suddenly with RedHat 8 there was a pre-set interface that knew much better what I wanted from a GUI. Or just not.
I had repressed that memory! Why did you have to bring it back?
Slashdot Mirror
I haven't even customized my user agent string and I'm using the standard Fedora 14 browser, but my user agent string itself is unique... Seems like I am the only Danish Fedora 14 user who has clicked on panopticlick recently.
One wonders what your position is regarding science and global warming.
Easy: the evidence indicates that our CO2 emissions are causing climate change. Caution should therefore lead us to reduce emissions until such time that we can be reasonably sure that we were wrong and CO2 emissions are safe.
Once we are reasonably sure that we can engineer GM crops to not spread their modified genes to other plants or become invasive species, we can reverse the ban. Assuming they do not compromise the security of our food sources in other ways -- e.g. monoculture is a risk whether it is GM or not.
If you do not donate all your money to the poor, you are in favour of letting humans be malnourished and starving.
But those people are in poor countries and have skin that is a different color than yours, so who really cares?
GM isn't the devil. It is just risky and the reward it offers is control of our food source handed to a select few.
So far mankind has proven itself spectacularly unable to handle risky technologies. The only real "success" so far is the ozone layer recovery.
Demand will be low. A War on GM plants would be over and done with in no time at all.
You will not have to do a field test if we succeed in getting GM crops banned.
Add a few tonnes of non-radiactive lead around the computer if you must. It is still much less expensive than the current neutrino detectors.
Yes, neutrinos are more common near nuke plants. At least that is what theory tells us. If you find a cheap way to PROVE this experimentally, you would become moderately famous among physicists. Getting extra glitches from memory would qualify...
You cannot encrypt a password both in transfer and on disk (unless you use a separately encrypted channel with separate authentication, but then why do you need a password?). For a lot of things it is more important that you can use the password to establish a secure channel than it is to store the password as a hash.
E.g. with the simple "ask for three specific characters from the password" method you gain almost-one-time-passwords, so a keylogger on a public terminal cannot empty your bank account afterwards. This property is probably worth the risk that someone breaks into the password storage.
So they're saying that enabling shader code execution allows web sites to exploit hypothetical vulnerabilities in the graphics driver?
They're not particularly hypothetical. Graphics driver code is such that games programmers carefully work around bugs in order to not crash anything. Imagine if every program running on the main CPU had to carefully avoid certain instruction sequences in order to not crash the system -- would you run a multi-user system on that?
Then again, that was how it was in the 80's on many time sharing systems...
It is probably not the cable provider doing it, but the content providers. They like to make demands that you only offer a particular channel to subscribers in a package with these other channels, or that you can only get this channel if you provide it to ALL your subscribers.
The currently standing record for GSM "decryption" is to pretend to be the provider and tell the phone to not encrypt. Sure, that does not involve any actual decryption, but since you as a user cannot tell it happened, what exactly is the difference?
You people keep repeating the same mistake over and over!
There is no problem sharing IP 2.2.2.2 port 34567 between thousands of connections. It is only a problem if all those users decide to connect to the same server on the same port simultaneously. This is highly unlikely, and few servers could handle that anyway. Google is NOT a counterexample, they use lots of IP addresses at their end.
No one does NAT based on just the 3-tuple, everyone does it on the 5-tuple. Except old Cisco products. 2^16 ports does not matter.
you do realize that that the port range is a 16-bit integer?
You do realize that proper NAT uses the whole 5-tuple for flow identification? The 16-bit port range is only a problem for old Cisco implementations of NAT; I am not aware of any non-Cisco devices with such a braindead design.
Are you aware that there is a limit on how many computers can be NATed behind a single IP address?
There may be a theoretical limit, but we are talking hundreds of thousands if not millions. In practice, you hit other limits first.
Nokia had smart phones basically since 9600bps became possible over GSM. They were heavy and awkward to use, of course, but the Windows CE phones were worse, so that is hardly an argument in favour of Microsoft.
The Icelandic government didn't borrow anything. The private bank Icesave borrowed a lot of money from banks and individuals primarily in the UK and the Netherlands. It went bankrupt, and the Dutch and UK governments decided to pay the banks and individuals what Icesave owed them. Very kind of those governments. Afterwards they decided they wanted the money from the Icelandic government.
Is it any surprise that the Icelandic public is failing to see why they should bail out private investors and banks in other countries, when no guarantee was issued in the first place? I am sure the US public is having similar difficulties with why they had to bail out the US credit institutions, but at least in that case they could vote out the politicians who did it.
An application could place its own window right at the place where I am expecting security-sensitive information to appear. Like right above the security lock icon in Firefox, making me believe that a particular site has a valid certificate. It is even worse if that affects focus, obviously.
And yes, X and the current window managers are broken in precisely this way right now. It does not matter too much because everyone implicitly trusts their X clients. In phone OS's, there is no such implicit trust, and we need to get there on the desktop too.
If you made your mirrors out of Faraday isolators, couldn't you theoretically have a mirror which admitted light from behind while reflecting light from the front? That ought to enable you to beat the concentration limit.
It is definitely not cost effective, of course.
Why not let your window manager do the job it was designed to do?
Because none of the common window managers actually do a decent job of managing windows. In turn, no one uses overlapping windows, instead switching between full-screen applications.
What really annoys me is that it is so trivially solved: the system should not raise (or lower or otherwise arrange, or map or unmap) a window EVER. The application must do it in response to events
That is completely the wrong way around. The application should stick to its own business and run completely sandboxed in its own window set, and let the user handle the placement and order. It is bloody annoying when Firefox manages to hijack the whole desktop and you cannot even switch virtual desktop. Applications should NOT be able to make modal dialog boxes. Allowing applications that kind of control is a security risk which is only acceptable because we have almost no sandboxing at all in all major desktop operating systems.
Phone OS's seem to be doing a lot better on the security side, and once the same kind of security is applied to the desktop, the windowing model will need to be secured. Phone OS's tend to handle the windowing model by not allowing overlapping windows at all.
You have never had to re-type something from one window to another, or simply had to react to something going on in one window by typing in another window? The first would be fixed by working copy-paste, but alas that seems to be getting increasingly broken even in Linux. Back in Red Hat 3 you could copy-paste all text except window titles; now some dialog boxes have text which cannot be copied.
I was totally happy with sawfish on 7.3, and suddenly with RedHat 8 there was a pre-set interface that knew much better what I wanted from a GUI. Or just not.
I had repressed that memory! Why did you have to bring it back?
In Windows, the active window must be on top, focus cannot follow mouse, there are no virtual desktops, and the application switcher is inconvenient.
To use Windows even somewhat effectively you need as many physical screens as you have active windows.