Whatever, I can "apt-get upgrade" Sid any time, no matter how long I've left the computer, and it works.
I do not really disagree. That's exactly my experience... Unless the computer has a NVidia GPU installed. When that's the case, you must carefully read the changes before applying, and consider your kernel something you don't upgrade.
Anyway, that's not a flaw of Debian. And the fact that you can upgrade Debian Stable at any time, even with an NVidia GPU is impressive... But Sid is not that stable.
Probably, most people here use Gnome, as it's been the default on most distros since the late 90's.
And every time Gnome 3 removes some functionality (altough it looks like this time it's optional), a few people cry about it, and a few of those change their DE. That's most of the noise you see here, the rest is people like me, pointing to them and saying "Ha Ha".
If it is indeed time to change (why change into GaAs? If you are changing, why not carbon?) you can expect that change to happen in a 15 to 25 years journey, as no fab is prepared for that, and no process works well on the next substrate yet.
RPi is cheap, and has all the I/O you'll ever want (for moderate values of "all"). Do not forget about the I/O, because it looks like that most board manufacturers do forget.
Well, that probably won't ever happen again, because this time we have vaccines. After a few months, the number of people that can have the flu will reduce fast.
But the next flu pandemics will spread faster too, making the vaccines way less usefull.
Please, tell me what Linux distro did require that since the last decade you: 1 - Buy newer versions. The problem here is money spending, so free updates won't do it. 2 - Change your GUI to something completely unrecognizable. Yeah, one of the 2 most used options from the last decade changed, and several people started using a third one, one of them is pretty like it was at 2001. And there are several GUIs intentionaly like the onld ones if you really need it. 3 - Change your computer more than once on the decade. 4 - Increase your workload in maintaining it. 5 - Use a more complex security system. 6 - Pay for SAAS.
Those are the issues the GP enumerated. Go ahead, name one distro that has them.
Few people ever tried to break your algorithm, thus it's untested and one shouldn't rely on it. Altough, it should be no worse than a single application of AES. Also, crypto researchers are always after fast algorithms that achieve as near as the maximum security as possible for their keylength, thus just doing more rounds won't gain many fans.
Now, I have a similar question. If I take 3 (or more) assymetrical crypto algorithms - name them A, B and C, and encrypt some text with them, in the order ABCABC, I should get something that is no worse than the best algorithm in the set. Is that a valid thing to do when you suspect one of them may have a backdoor?
Backdooring a network card would be easier (it has first-hand access to the network data), more transparent, less obvious and as usefull than backdooring a processor.
It's quite harder to see anything wrong when two devices talk to each other by DMA, completely bypassing anithing that could run an OS.
Well, by the other side, Al Qaeda people are probably smart enough to use a longer key, so all the NSA will have the power to breeak are the communication of people like you.
And if they ever spend that billion to acquare the capacity, they'll use it, whatever targets they have.
He said exactly that some ECC curves have suspicious origins, and one shouldn't trust them. What he didn't say is that all of ECC is suspicious, or that he knows anything with certainty.
Anyway, it's easier to use RSA with a larger key than to investigate each ECC curve you are thinking about using. But that's just my opinion.
Slashdot Mirror
Have you tried installing another DE? (I recommend KDE.) You know, whatever distro you are using, Unity is not mandatory.
I do not really disagree. That's exactly my experience... Unless the computer has a NVidia GPU installed. When that's the case, you must carefully read the changes before applying, and consider your kernel something you don't upgrade.
Anyway, that's not a flaw of Debian. And the fact that you can upgrade Debian Stable at any time, even with an NVidia GPU is impressive... But Sid is not that stable.
Probably, most people here use Gnome, as it's been the default on most distros since the late 90's.
And every time Gnome 3 removes some functionality (altough it looks like this time it's optional), a few people cry about it, and a few of those change their DE. That's most of the noise you see here, the rest is people like me, pointing to them and saying "Ha Ha".
The Reprap Mendel is about $600 nowadays, but unassembled.
Nah... I'm pretty sure tath in a few years nobody would complain anymore.
If it is indeed time to change (why change into GaAs? If you are changing, why not carbon?) you can expect that change to happen in a 15 to 25 years journey, as no fab is prepared for that, and no process works well on the next substrate yet.
One really shouldn't hold one's breath.
It's $25/$35.
RPi is cheap, and has all the I/O you'll ever want (for moderate values of "all"). Do not forget about the I/O, because it looks like that most board manufacturers do forget.
Ok, than, woosh for me.
Well, that probably won't ever happen again, because this time we have vaccines. After a few months, the number of people that can have the flu will reduce fast.
But the next flu pandemics will spread faster too, making the vaccines way less usefull.
That's caled television.
I guess we can skip that "virus" part, telecontol is a builtin feature of humans.
Please, tell me what Linux distro did require that since the last decade you:
1 - Buy newer versions. The problem here is money spending, so free updates won't do it.
2 - Change your GUI to something completely unrecognizable. Yeah, one of the 2 most used options from the last decade changed, and several people started using a third one, one of them is pretty like it was at 2001. And there are several GUIs intentionaly like the onld ones if you really need it.
3 - Change your computer more than once on the decade.
4 - Increase your workload in maintaining it.
5 - Use a more complex security system.
6 - Pay for SAAS.
Those are the issues the GP enumerated. Go ahead, name one distro that has them.
Very true. And all theories about the real world are either false or undecidable.
Now, at the real world, we like to qualify that undecidable set.
That number is called "googol", not "google".
They can't afford a slightly more expensive option, but can afford relplacing it every 5 years?
TFS says they tested that in yeast, thus you either didn't care enough to read it, or is asking if the yeast mental activity was degraded.
You can discover that easily if you can read the network stream.
Few people ever tried to break your algorithm, thus it's untested and one shouldn't rely on it. Altough, it should be no worse than a single application of AES. Also, crypto researchers are always after fast algorithms that achieve as near as the maximum security as possible for their keylength, thus just doing more rounds won't gain many fans.
Now, I have a similar question. If I take 3 (or more) assymetrical crypto algorithms - name them A, B and C, and encrypt some text with them, in the order ABCABC, I should get something that is no worse than the best algorithm in the set. Is that a valid thing to do when you suspect one of them may have a backdoor?
Backdooring a network card would be easier (it has first-hand access to the network data), more transparent, less obvious and as usefull than backdooring a processor.
It's quite harder to see anything wrong when two devices talk to each other by DMA, completely bypassing anithing that could run an OS.
All of Intel last processors come with builtin RNGs... And it seems that they have backdoors from the NSA.
Don't trust your hardware too much.
Well, whatever is the target (even Washington), the end result is the same.
Well, by the other side, Al Qaeda people are probably smart enough to use a longer key, so all the NSA will have the power to breeak are the communication of people like you.
And if they ever spend that billion to acquare the capacity, they'll use it, whatever targets they have.
Also, whatch your random number generator. Once you start with such schemes, it starts to become a bottleneck.
He said exactly that some ECC curves have suspicious origins, and one shouldn't trust them. What he didn't say is that all of ECC is suspicious, or that he knows anything with certainty.
Anyway, it's easier to use RSA with a larger key than to investigate each ECC curve you are thinking about using. But that's just my opinion.
Well, come back in 4 years. We'll see.