The Good & Bad Of ZFS + HAMMER File-Systems On BSD

[dante]

They both suck and are in no way represent the future of file systems. Want to see the future of file systems? Look at btrfs and ext4 (especially btrfs).

By the way, it's meaningless to just compare ZFS and HammerFS as they have the same shit design, same shitty design team mentality and aptitude and are almost as worse as each other (HammerFS being slightly worse). To make sense, we must include ZFS, EXT4FS and Resier4. ZFS feature's aren't Sun's invention. They were copied from Hans Resier's design together with some tweets and a whole lot of flaws due to the bureaucracy at Sun. HammerFS is nothing but a BTRFS/ZFS wannabe, unfortunately mad dog Dillion didn't have to skill or resources to candy coat shit. BTRFS takes all the good parts from ZFS and implements them correctly together with lots of innovative features from oracle.

I`m not a huge fan of ZFS but BTRFS is clearly inspired on ZFS maybe will be just as good or better but is still an experimental file system not production ready

[TeamBlackFox]

I agree with most of what you wrote, but please don't make things up, yourself. systemd does a lot of stuff, however, it does not tie the device manager (udev) to the init system. Udev is developed in the very same repository, it imports a few sd headers on compile time, however there is absolutely no runtime dependency! Gentoo and Debian - just to name two - compile udev from systemd sources and use it without systemd as pid 1.

Probably the cause of bad word choice here - I was actually referring to the fact that it is in the same project, and thus far I've not seen anyone use systemd with any other device manager system. Of course I know udev can be run without systemd, but not the other way around! I may not see uselessd, and the systemd-init as the ideal init system, but if it had stayed as that, and that alone, I'd be less opposed to its existence - competition with init is fine, but it went above and beyond and took it too far.

[mpppp]

Can't you see. It is that fragmentation that prevents Linux to take the place of that walled garden desktop OSes

Yes, everybody keeps repeating that. But maybe it's the absence of walls instead.

Who uses linux on the desktops? Public administrations because of CONTROL (either they want to stop surrendering control to microsoft and its licenses, or, for corrupt places, they want an excuse to spend money on new contracts and will keep switching back and forth). Valve because of CONTROL (they saw Microsoft reimplementing steam, remembered the basically empty list of MS partners that didn't get screwed in the end and played the let's own the OS card). Geeks and hobbyists because they like or need CONTROL.

You might say "but if they want CONTROL why not use linux, since you can actually own it and control it?"
The reason everybody is not jumping ship and own the OS, is that proprietary or rapidly moving stacks (android, and if it's not too early to say, systemd) make it easier for hardware makers and software makers to have:
- planned obsolescence thanks to incompatibilities
- binary only , or protected/DRMd software distribution.