Because not only the code would need to be portable, but the build system too (makefiles). Also, in 25 years you don't want to re-qualify everything and risk introducing some new bugs because the new compiler doesn't behave like the old one. Too risky.
According to Slashdot it affected only kernels 3.4 to 3.6. I wasn't using any of those kernels so I was fine. It's more an argument against those specific (unpatched) kernel versions than against ext4 as a filesystem.
Every kernel module not in the mainline kernel suck compared to an integrated one. It needs to be recompiled for every new kernel, often breaking things, and in many cases (such as this one), taints the kernel so you loose a lot of support.
I know, but language isn't the same as math. Three times faster means 300%. Just like three times slower means one third, not that you are going faster but in the opposite direction.
Actually, I understood it the other way. 2x faster means twice as fast, 100% faster, or 200% of the baseline. 1.4x faster is the same as 40% faster, 140% of the baseline.
ZFS is GPL-incompatible, it will never reach wide acceptance in the Linux* community (or Windows). Unless your PC has a single purpose of storing files, I doubt the file system is an important enough reason to switch to another OS. BTRFS has similar functionality and is becoming mature quickly, so that will probably be my next file system if I ever need these features or when it becomes the default in Linux distributions.
*Yes, I know there is a fuse project and a 3rd party ZFS kernel module but they both suck compared to a real, integrated file system driver.
It didn't happen to me. I wasn't even aware there was such a large-scale bug. And it's not as if reseirfs or XFS were bug-free either. If I were scared by ext4, I'd use ext3, not reiser or XFS.
student gets higher interest rates on loans and refused credit. It may sounds "brilliant" at first, but if everyone were doing the same, it wouldn't work. This guy should thanks all others who will end-up paying for him.
That there is a cost to transition from imperial to metric isn't backed by evidence (at least, not in this conversation) and is definately not an opinion.
The hiatus wasn't a stop of global warming. It was a decrease of the speed of the warming. Therefore even if the hiatus is real, it doesn't mean global warming has stopped, unlike what deniers have been saying.
Slashdot Mirror
If they wanted standard cables and peripherals, they wouldn't be buying Apple stuff.
SM what?
Using new compilers was the whole point of the OP. Of course if you keep the old compiler you can keep the old "make" utility too.
Because not only the code would need to be portable, but the build system too (makefiles). Also, in 25 years you don't want to re-qualify everything and risk introducing some new bugs because the new compiler doesn't behave like the old one. Too risky.
Yeah, use something already outdated to make sure that no one evens remember it in 25 years.
One more reason to browse using Firefox which allows ad-blockers (at least on Android), unlike stock Android or iOS browsers.
It varies wildly between phone manufacturers and it is dominated by 2 or 3 companies. Don't go down that path.
BB10 varies even more from stock Android, so what is your point?
Of course I understand the difference, however, call the article wrong if you like, but I am 99% sure that the author meant 40% faster.
I understand most people are wrong, but at least I understand what they mean. The purpose of language is to be understood by others.
I understand all that. It's just not what people say.
According to Slashdot it affected only kernels 3.4 to 3.6. I wasn't using any of those kernels so I was fine. It's more an argument against those specific (unpatched) kernel versions than against ext4 as a filesystem.
Every kernel module not in the mainline kernel suck compared to an integrated one. It needs to be recompiled for every new kernel, often breaking things, and in many cases (such as this one), taints the kernel so you loose a lot of support.
who was affected? A bug that affects only 0.0001% of users is a lot less relevant.
40% faster is not slower. But nobody will say 0.4x faster. 0.4x means slower, yes.
I know, but language isn't the same as math. Three times faster means 300%. Just like three times slower means one third, not that you are going faster but in the opposite direction.
Actually, I understood it the other way. 2x faster means twice as fast, 100% faster, or 200% of the baseline. 1.4x faster is the same as 40% faster, 140% of the baseline.
ZFS is GPL-incompatible, it will never reach wide acceptance in the Linux* community (or Windows). Unless your PC has a single purpose of storing files, I doubt the file system is an important enough reason to switch to another OS. BTRFS has similar functionality and is becoming mature quickly, so that will probably be my next file system if I ever need these features or when it becomes the default in Linux distributions.
*Yes, I know there is a fuse project and a 3rd party ZFS kernel module but they both suck compared to a real, integrated file system driver.
- Lenders may be asking more questions of students - job prospects, ability to pay, blah blah.
They already do.
It didn't happen to me. I wasn't even aware there was such a large-scale bug. And it's not as if reseirfs or XFS were bug-free either. If I were scared by ext4, I'd use ext3, not reiser or XFS.
student gets higher interest rates on loans and refused credit. It may sounds "brilliant" at first, but if everyone were doing the same, it wouldn't work. This guy should thanks all others who will end-up paying for him.
ext4 was more than ready when it became default in Debian.
That there is a cost to transition from imperial to metric isn't backed by evidence (at least, not in this conversation) and is definately not an opinion.
True, but AGW is already confirmed (with a >95% confidence), hiatus or not.
The hiatus wasn't a stop of global warming. It was a decrease of the speed of the warming. Therefore even if the hiatus is real, it doesn't mean global warming has stopped, unlike what deniers have been saying.
That alone is worth whatever price is may or may not "cost" us
Do you have any evidence about that?