Reiser4 May Go For Mainline Inclusion In 2010

[phoronix]

Phoronix: Reiser4 May Go For Mainline Inclusion In 2010

The Reiser4 file-system has been around since 2004 but has not reached a point of being close to be included in the mainline Linux kernel, especially after the lead developer, Hans Reiser, was convicted of murdering his wife. Development of Reiser4 has continued on, albeit with a very limited number of developers, and not nearly at the brisk pace of Btrfs or with great interest by corporate parties...

http://www.phoronix.com/vr.php?view=NzY4OQ

[DarkCloud]

Can the name somehow get changed, seems in appropriate to have a name of a file system after a convicted murder. It would be nice to honour the woman killed by this asshole and name it after Reiser's deceased wife

[rohcQaH]

Originally Posted by DarkCloud

name it after Reiser's deceased wife

coincidentally, her name was Reiser too. (duh!)

[Danyul]

She did take his name when she married him so that would name it...uhhh....Reiser?! (I'm thinking you probably were thinking maiden name and Sharanova would make an interesting name for an FS)

But hey, there are plenty of things named after people much worse than him. One thing off the top of my head is Columbus Day. I'm sure the other bright sparks here could name a few. (Still too early for me)

I personally don't care what it's called. If it's good (and outlook is positive) and included in the kernel, I will use it.

[Fixxer_Linux]

Beyond the name of this filesystem, I guess that the main challenge is to maintain the code along kernel updates even if than being able of hitting the kernel is already a challenge.

Therefore, who will take the risk to switch to Reiser4 if there is not many people to maintain it ?!? (just guessing though, as I'm no dev at all).

Wouldn't it be more usefull that the people involved in Reiser4 could involve themselves in btrfs instead (and including all the Reiser4 code that can be included in btrfs or ext4) ??

[Ex-Cyber]

Originally Posted by Fixxer_Linux

Wouldn't it be more usefull that the people involved in Reiser4 could involve themselves in btrfs instead (and including all the Reiser4 code that can be included in btrfs or ext4) ??

It doesn't really work that way with software projects in general, let alone filesystems. Typically, the reason there's more than one project in the first place is that there's more than one idea about how things should be done.

[Adarion]

1st
Whoa. Hell finally froze over. Right?
Can't believe it. How old is Reiser4 now?
Anyway good to hear that this much praised and perfomant FS will finally enter the Kernel.

2nd
Who wants to throw the first stone? Do you folks here really know all the details what happened in the Reiser case? Does anybody know the truth? No?
Then don't throw your stones.
It's a longer story between both of them as far as I see it.
So a rename might be okay but then it should be free of any Reisers. No Hans or Nina. Maybe give it just a neutral name.

[charlie]

I hereby call it, "Atomic File System". Notify the rest of the Linux world.

[Jimmy]

Originally Posted by Adarion

Do you folks here really know all the details what happened in the Reiser case?

1. He was convicted. 2. He confessed. 3. He led authorities to the body. But I'm sure he didn't do anything after all I don't know what color of socks he had on.

I think we know enough. I don't want to know any more details. The sick fuck head killed his wife and no matter how much of a genius file system hero he may be to some I really think people need to come to grips with reality. Stop going to conspiracy sites. Mankind did walk on the moon, 9/11 wasn't carried out by the CIA, and Hans Riser did kill his wife. Get over it.

[kernelOfTruth]

definitely great news !

kudos to Edward and the other developers that drive forward this great filesystem

there are some fundamentals / great ideas that got implemented in Reiser4:

thus some proposals for a name:

* dancing (tree) filesystem
* atomic filesystem
* linux fibration filesystem
* tailpacking filesystem2
* ...

IMO main advantages are:

- dancing trees / "trees that cache well"
- transparent compression
- efficiency with small and lots of files
- performance / parallelism
- easy on hardware (especially harddrives) - compare it with ext3 or xfs and you'll notice that harddrives work much quieter (they don't "trash" that much / don't seem to have to move their heads that often)
- and other nice functionality -> read the code