The world as a whole. I have far more confidence in 50% of the world remaining free than I do in the US remaining free, and the fact that you have to make such bogus arguments (yeah, china's really representative of the UN as a whole) suggests to me that the US doesn't have a real reason and is just saying "mine, my preciousss".
Look also at the ability to obtain a list of every base station that your mobile is associated with - the phone companies can do this if requested by a magistrate, although that usually only done in murder cases or similar.
That's qualitatively different. Once you have a warrant, knock yourself out - but I'm not happy about people being routinely tracked.
In such situations, CRC32 is far simpler and just as effective. MD5 was designed as a cryptographic hash and isn't really good for anything else. Now we see it's no good as a cryptographic hash either. It's basically no good.
Can someone explain to me who would trust the security of software written in C?
Languages don't make things secure, good programming makes things secure. PGP, GPG, OpenSSH and OpenSSL were written in C last time I checked.
If you want something to be portable and usable from different languages, which the base protocol needs to be, you need to be able to make libraries with C linkage - and C is really the only choice for such things unless you have a specialised team who are good at something like Delphi. Well written code is still secure, and the gnunet code is some of the best code I have ever worked with, certainly far clearer than other p2p code e.g. gift.
Gnunet is here and working. Fully usable as a P2P network, not as fast as unencrypted but close. I haven't tried using it in pure friend-to-friend mode but the functionality is there. And of course it has all the things you'd expect from an advanced P2P network, searches for automatically extracted keywords, signed namespaces where you can publish content anonymously but show that it's all from you, directories, etc.
That's simply not true. Early 56ks would drop down to 33.6 at the drop of a hat, but if you had a good line it was possible to get 56 both ways to another modem the same.
No, if it's google then slashdot loves them, this trumps all other considerations. (Except possibly apple. I wonder what slashdot would do if google and apple went against each other)
If people *can* steal something, they will. The honor system doesn't work.
The fact that people are still buying CDs proves otherwise. None of this DRM stuff actually stops anyone copying it. If you want to download the music, hop on any P2P network around and grab a copy, these days you can usually even find a flac version so it will be just as good as the CD. But people do still buy the CD.
I don't see how that one won, it's not a valid C program. It won't compile without -nostdlib, and if you do it segfaults when you try and run it, meaning you can't just feed its output into a c compiler and get another copy of it as you would expect. It's a clever idea and would work in, say, perl, but did they even test it?
Programmers can introduce security vulnerabilities in any language. Bad programming is not language specific.
True, however some languages are easier to introduce flaws in than others - when was the last time you saw a buffer overflow in a modern language? They do happen but only occasionally. PHP programs seem to have flaws far more often than others.
If the kernel developers aren't willing to provide a stable interface then they should be more ready to accept code into the kernel. Too many things (in particular I'm thinking of reiser4 here) are kept out for political reasons. I can understand not wanting to make a stable interface, but in that case you have a responsibility to accept code that works and is GPL without making people jump through hoops.
The crucial difference here is that companies can ship commercial software on top of a Linux kernel without their software falling under the GPL and without paying anybody
No, but then they aren't allowed to use lots of the nice things in the kernel and may have to reimplement lots of functionality that they could just get immediately if they were GPL.
while companies cannot ship commercial software on top of Qt without their software falling under the GPL or paying a lot of money to TT. And you can argue that that doesn't matter, but it obviously does.
Doesn't seem so obvious to me. If it's that much trouble they can always use another toolkit - their program will still work in KDE.
Gnome is as relevant as it has always been,
The stated aim of the gnome project is to ensure there's a Free Software desktop that's as good as KDE. It is now irrelevant to that original intention.
and Novell's decision underlines that again: KDE can never become the standard desktop for Linux. It's just not going to happen.
I could say the exact same of gnome. In a sense you're right - KDE would never become "the standard", because KDE devs believe in choice and would always make sure you can use the important bits of KDE without having to run KDE. But in terms of being the desktop that everyone uses, KDE could certainly do it.
Slashdot Mirror
Just sayin'
The world as a whole. I have far more confidence in 50% of the world remaining free than I do in the US remaining free, and the fact that you have to make such bogus arguments (yeah, china's really representative of the UN as a whole) suggests to me that the US doesn't have a real reason and is just saying "mine, my preciousss".
Which is absolutely how it should be. There are plenty of examples of bad laws by our government, but this is not one of them.
That's qualitatively different. Once you have a warrant, knock yourself out - but I'm not happy about people being routinely tracked.
No, they are very high. Almost a certainty in fact.
In such situations, CRC32 is far simpler and just as effective. MD5 was designed as a cryptographic hash and isn't really good for anything else. Now we see it's no good as a cryptographic hash either. It's basically no good.
Verify the pgp signatures instead, and check they were done with a better algorithm
Oh, I have the money, I just need to move to antarctica first.
No, multiple hash routines *does not work*, as is said in every single hash related thread. Fortunately I'm pretty sure BT uses a newer hash already.
Languages don't make things secure, good programming makes things secure. PGP, GPG, OpenSSH and OpenSSL were written in C last time I checked.
If you want something to be portable and usable from different languages, which the base protocol needs to be, you need to be able to make libraries with C linkage - and C is really the only choice for such things unless you have a specialised team who are good at something like Delphi. Well written code is still secure, and the gnunet code is some of the best code I have ever worked with, certainly far clearer than other p2p code e.g. gift.
Gnunet is here and working. Fully usable as a P2P network, not as fast as unencrypted but close. I haven't tried using it in pure friend-to-friend mode but the functionality is there. And of course it has all the things you'd expect from an advanced P2P network, searches for automatically extracted keywords, signed namespaces where you can publish content anonymously but show that it's all from you, directories, etc.
My dad listens to podcasts. They're not a geek thing anymore. Remember how filesharing programs didn't used to have ads?
That's simply not true. Early 56ks would drop down to 33.6 at the drop of a hat, but if you had a good line it was possible to get 56 both ways to another modem the same.
No, if it's google then slashdot loves them, this trumps all other considerations. (Except possibly apple. I wonder what slashdot would do if google and apple went against each other)
There's no way it was being said sarcastically, nor have I ever seen it used as such. The book's got it wrong.
The phrase is "couldn't care less", only idiots who've misheard say "could care less". Ignore them.
The fact that people are still buying CDs proves otherwise. None of this DRM stuff actually stops anyone copying it. If you want to download the music, hop on any P2P network around and grab a copy, these days you can usually even find a flac version so it will be just as good as the CD. But people do still buy the CD.
I don't see how that one won, it's not a valid C program. It won't compile without -nostdlib, and if you do it segfaults when you try and run it, meaning you can't just feed its output into a c compiler and get another copy of it as you would expect. It's a clever idea and would work in, say, perl, but did they even test it?
True, however some languages are easier to introduce flaws in than others - when was the last time you saw a buffer overflow in a modern language? They do happen but only occasionally. PHP programs seem to have flaws far more often than others.
If the kernel developers aren't willing to provide a stable interface then they should be more ready to accept code into the kernel. Too many things (in particular I'm thinking of reiser4 here) are kept out for political reasons. I can understand not wanting to make a stable interface, but in that case you have a responsibility to accept code that works and is GPL without making people jump through hoops.
Who's laws are you going to be subject to? The country you left from? A new lunar republic?
Hope claims that while it is illegal for countries to end sentences with prepositions, it is legal for individuals and corporations to.
The best windows one, KCeasy, got takedown'd and made to stop distributing the fasttrack plugin.
No, but then they aren't allowed to use lots of the nice things in the kernel and may have to reimplement lots of functionality that they could just get immediately if they were GPL.
while companies cannot ship commercial software on top of Qt without their software falling under the GPL or paying a lot of money to TT. And you can argue that that doesn't matter, but it obviously does.
Doesn't seem so obvious to me. If it's that much trouble they can always use another toolkit - their program will still work in KDE.
Gnome is as relevant as it has always been,
The stated aim of the gnome project is to ensure there's a Free Software desktop that's as good as KDE. It is now irrelevant to that original intention.
and Novell's decision underlines that again: KDE can never become the standard desktop for Linux. It's just not going to happen.
I could say the exact same of gnome. In a sense you're right - KDE would never become "the standard", because KDE devs believe in choice and would always make sure you can use the important bits of KDE without having to run KDE. But in terms of being the desktop that everyone uses, KDE could certainly do it.
Protocols and services are pretty much the same thing as far as the internet's concerned.