Yeah, YOU will, but too many people aren't security-conscious enough. People don't check return codes where they should (close() is a good one), and they spend too much time hacking and not enough time planning and keeping the main design simple (see Gnome/KDE).
Eventually, we will all use some version of these new ideas, because as you may have noticed, dividing processes into root/user just leads to root exploit after root exploit.
Sigh. Something that wouldn't happen if people would pay attention when they designed and wrote their root daemons. Qmail's never been cracked like that (No, I'm not saying I like DJB, but his code is pretty good.)
Actually, they are. Linus moved to the U.S. a while ago. (Though I agree with Alan Cox that it's probably dangerous for developers to visit the U.S., due to the crazy laws.)
I hope you can come up with enough of your own U.S.-born tech workers, because I sure as hell won't be moving to the U.S. any time soon (and I'll leave Canada if it gets bad here, too. I have no loyalty to a country that fscks around its citizens.)
Here's the problem with that: you won't really *know* whether you have good security or not unless there is some sort of peer review. Otherwise, you may have missed something that is easy to exploit.
Um... Hardware costs money, and even if it didn't, it would be much better to have efficient software that screams on fast hardware than to have bloated software that runs O.K. on the same hardware.
I have a Thunderbird 1.2GHz system with 512MB SDRAM, and I still run WindowMaker+fspanel, which runs adequately on a P120.
It's not that bloated, you just have to know what software to install (though I admit FreeBSD is much leaner, anyway).
I run WindowMaker with fspanel on my Thunderbird 1.2GHz, and the system just flies. What's interesting is that you can run the same configuration on a P120 and it's still very usable. I can't say the same for GNOME or KDE.
In other words, Linux-based systems aren't really bloated, unless you want that spiffy new eye-candy-filled desktop environment.
Let's not forget that this fight for strong cryptography isn't just a fight for privacy, but for network security on the internet. Strong cryptography is the key cornerstone of internet security, and hindering it in any way will result in the compromise of those systems.
Internet security is based on a trust/no-trust system, and the one common trusted thing among all security protocols is the security of the crypto. If this trust system is undermined, as will be the case with restricted and/or "backdoored" crypto, then the entire trust system collapses. We have to "hope" our systems will remain secure, and we can no longer trust that they are.
Despite my bad English, what I'm trying to say is that key escrow, backdoors, and other similar man-made vulnerabilities in crypto will disrupt the functioning of the internet and e-commerce more than most people think. So, anyway, this isn't just a battle to be fought by "liberal-minded fools crying for rights that don't really exist", this is a battle for internet security which needs to be fought by everyone.
This isn't really about copyright, it's about ethics. Even without copyright, it would still be wrong to take someone else's work and claim it was your own.
Slashdot Mirror
The Church of Emacs is not to be confused with the GPL.
What a way to ruin your child's future!
Hah! The foil does nothing!
Yeah, YOU will, but too many people aren't security-conscious enough. People don't check return codes where they should (close() is a good one), and they spend too much time hacking and not enough time planning and keeping the main design simple (see Gnome/KDE).
Dwonis ducks.
Sigh. Something that wouldn't happen if people would pay attention when they designed and wrote their root daemons. Qmail's never been cracked like that (No, I'm not saying I like DJB, but his code is pretty good.)
Actually, they are. Linus moved to the U.S. a while ago. (Though I agree with Alan Cox that it's probably dangerous for developers to visit the U.S., due to the crazy laws.)
I hope you can come up with enough of your own U.S.-born tech workers, because I sure as hell won't be moving to the U.S. any time soon (and I'll leave Canada if it gets bad here, too. I have no loyalty to a country that fscks around its citizens.)
Yes, this is "more secure", but not "unbreakable". There's a big difference. One is theoretically possible to break/guess, the other is not.
"But why should I have to pay for it??"
Because my colleague at work need to send me a new Kerberos key when my laptop gets compromised.
Like my sig says, internet security depends on strong cryptography. Full stop.
Yes, but this changes the status of the algorithm from unbreakable to not feasible to break.
The MD5 is:
3ca9f611a036243a24e582b24bf127a8 linuxq3apoint-1.30.x86.run
Anyone want to check if this is the same as the official one?
Well if ISPs would start supporting multicast (or IPv6!), this wouldn't be consuming massive amounts of bandwidth.
</rant>
Here's the problem with that: you won't really *know* whether you have good security or not unless there is some sort of peer review. Otherwise, you may have missed something that is easy to exploit.
Especially once you've installed Debian! Nearly everything you ever wanted is just an apt-get away!
Didn't MSOffice2000 switch to an XML format?
I have a Thunderbird 1.2GHz system with 512MB SDRAM, and I still run WindowMaker+fspanel, which runs adequately on a P120.
Telnet? Are you nuts?? SSH with PublicKey authentication.
I run WindowMaker with fspanel on my Thunderbird 1.2GHz, and the system just flies. What's interesting is that you can run the same configuration on a P120 and it's still very usable. I can't say the same for GNOME or KDE.
In other words, Linux-based systems aren't really bloated, unless you want that spiffy new eye-candy-filled desktop environment.
Yes, there are uses for Apache on Windows -- all I said was there wouldn't be many people doing that.
Internet security is based on a trust/no-trust system, and the one common trusted thing among all security protocols is the security of the crypto. If this trust system is undermined, as will be the case with restricted and/or "backdoored" crypto, then the entire trust system collapses. We have to "hope" our systems will remain secure, and we can no longer trust that they are.
Despite my bad English, what I'm trying to say is that key escrow, backdoors, and other similar man-made vulnerabilities in crypto will disrupt the functioning of the internet and e-commerce more than most people think. So, anyway, this isn't just a battle to be fought by "liberal-minded fools crying for rights that don't really exist", this is a battle for internet security which needs to be fought by everyone.
OTOH, virus writers do a great job of refuting people's claims of security, and making those refutations public... Go figure.
Not many. Most people clueful enough to use apache wouldn't waste it on such a crappy OS.
This isn't really about copyright, it's about ethics. Even without copyright, it would still be wrong to take someone else's work and claim it was your own.