Re: Overwrite existing partition with zeros without hurting partition table? (Debian Lenny)



disclaimer: I have no interest in the future of reiserfs3 or
4 and I've never been a user of it (although I have tried it
out once or twice). I personally feel that any concepts
embedded within reiserfs that are worth salvaging would be
better implemented in another code-base.

On Sun, Mar 07, 2010 at 11:21:19PM -0600, Stan Hoeppner
wrote:
For an educated man you make many assumptions, many
incorrectly. I didn't say the code was bad because Reiser
is a bad man. I said one of the reasons I wouldn't use
ReiserFS is because its author is a convicted murderer.
This has nothing to do with code, but supporting the work
of a criminal.

"ReiserFS" might mean ReiserFS3 (in-kernel) or ReiserFS4
(not).

The copyright holders for ReiserFS3, as represented
in-kernel, are mixed, but largely Hans Reiser via copyright
assignment. The authors of the code are mixed, and it's
harder to say who wrote what, but I think the majority was
not by Reiser himself. (A list of the main contributors is
present in fs/reiserfs/README in the Linux source tree)

The copyright holder for all code in ReiserFS4 is namesys,
which is (or was) Hans Reiser (again due to copyright
assignment). Again it's hard to say who "wrote" ReiserFS4.
Vladimir Saveliev is described as the "main programmer".
There are at least seven other major contributors.

It's akin to not eating at a known Mafia owned/operated
restaurant. The food is most likely fantastic, but I
wouldn't eat there because of who owns the joint. I don't
condone or support criminality.

You are of course entitled to your own moral code and
values. This comparison suffers on one front, though:
eating in a mafia controlled restaurant means you are
directly financing crime.

It would appear Reiser's crime is entirely unrelated to the
ReiserFS filesystem family. Using reiserfs3 or 4 less
obviously or directly supports Reiser in any fashion.

There's a possibility that using reiserfs4 contributes
towards it's popularity, which in turn might have an
influence on the likelyhood of commercial contracts with
Namesys for developing reiserfs4, which would result in
financially supporting Reiser. But the company is currently
suspended and unable to take on contracts and it would seem
vanishingly unlikely that it will return.

Novell, being headquartered in the Mormon capital of the
world, Salt Lake City, Utah, announced the day Reiser was
arrested (not convicted but arrested) that they were
dropping ReiserFS as the default filesystem in SLED/SLES
immediately, and would be moving to help permanently phase
it out at customer sites ASAP. Obviously this latter part
didn't happen over night as it was deployed on many
production servers. Novell didn't do this because the
code was bad. They did it because religious/political
forces dictated they distance themselves asap from Hans
Reiser, especially given competent (and some would argue
better) alternatives were available. We haven't even
touched the practical issues involved in continuing use of
ReiserFS.

Some might argue that they needed an exit strategy from
ReiserFS (since the likelyhood of v4 being merged was
vanishingly small, and v3 was largely unmaintained), and
Reiser's conviction gave them just that.


--
Jon Dowland

Attachment: signature.asc
Description: Digital signature