3 ms·

Wouldn't it be cool if Apple just picked up a standard file system from somewhere else? Ext4 for example.

coldtea 11y ago
They would still keep something like resource forks. Most modern filesystems have them. Here's the man page for Ext4 and co on them: http://manpages.ubuntu.com/manpages/precise/man5/attr.5.html http://manpages.ubuntu.com/manpages/precise/man5/attr.5.html
- duskwuff 11y ago
  Heck - at this point, resource forks are exposed as a "com.apple.ResourceFork" xattr with a large binary value. The primary remaining visible quirk is that they're also accessible via a magic path ("file/..namedFork/rsrc") -- the fact that the file system has a special way of storing them is irrelevent to userspace.
  - yrro 11y ago
    Ah, that explains why people in this thread are so confused between alternate forks/streams, and extended attributes!
mirekrusin 11y ago
ext4 deprecation hangs in the air and it's not very good filesystem, btrfs is probably the future and it's slowly becoming production ready.
- justizin 11y ago
  yah absolutely ext4 has progressively had its' performance reduced because what made it faster than ext3 created a lot of corner cases that make it far less reliable, and Btrfs is at least a fresh design, but doesn't seem to be particularly ready and is not even proven as a reliable design. It's good that we have more than one filesystem. HFS+ functionally works well for a Mac because the Mac is designed around it, but it is part of the reason that OSX falls short as a UNIX at times.
  - Skunkleton 11y ago
    Yes it is likely that ext4's days a numbered in terms of developing new features, however _today_ it is still the best choice for a filesystem if you don't run a bleeding edge kernel, and don't need the extra features. BTRFS is seen as the next defacto standard linux filesystem, and it has a ways to go before it can match the performance of ext4.
  - the_why_of_y 11y ago
    > what made it faster than ext3 created a lot of corner cases that make it far less reliable That is an odd claim, considering that the ext3 implementation has been removed from the kernel this year, and various distributions have for several years already used the ext4 driver to mount ext3 filesystems. https://lkml.org/lkml/2015/8/31/22 https://lkml.org/lkml/2015/8/31/22
bluedino 11y ago
There's that whole backwards compatibility thing.

They would still keep something like resource forks. Most modern filesystems have them. Here's the man page for Ext4 and co on them: http://manpages.ubuntu.com/manpages/precise/man5/attr.5.html http://manpages.ubuntu.com/manpages/precise/man5/attr.5.html

duskwuff 11y ago
Heck - at this point, resource forks are exposed as a "com.apple.ResourceFork" xattr with a large binary value. The primary remaining visible quirk is that they're also accessible via a magic path ("file/..namedFork/rsrc") -- the fact that the file system has a special way of storing them is irrelevent to userspace.
- yrro 11y ago
  Ah, that explains why people in this thread are so confused between alternate forks/streams, and extended attributes!

mirekrusin 11y ago

ext4 deprecation hangs in the air and it's not very good filesystem, btrfs is probably the future and it's slowly becoming production ready.

justizin 11y ago
yah absolutely ext4 has progressively had its' performance reduced because what made it faster than ext3 created a lot of corner cases that make it far less reliable, and Btrfs is at least a fresh design, but doesn't seem to be particularly ready and is not even proven as a reliable design. It's good that we have more than one filesystem. HFS+ functionally works well for a Mac because the Mac is designed around it, but it is part of the reason that OSX falls short as a UNIX at times.
- Skunkleton 11y ago
  Yes it is likely that ext4's days a numbered in terms of developing new features, however _today_ it is still the best choice for a filesystem if you don't run a bleeding edge kernel, and don't need the extra features. BTRFS is seen as the next defacto standard linux filesystem, and it has a ways to go before it can match the performance of ext4.
- the_why_of_y 11y ago
  > what made it faster than ext3 created a lot of corner cases that make it far less reliable That is an odd claim, considering that the ext3 implementation has been removed from the kernel this year, and various distributions have for several years already used the ext4 driver to mount ext3 filesystems. https://lkml.org/lkml/2015/8/31/22 https://lkml.org/lkml/2015/8/31/22

bluedino 11y ago

There's that whole backwards compatibility thing.