Bad, vile and meaningless: Hans Reiser is an asshat from Alan's clob

Hans Reiser is an asshat

What is an asshat? Don't ask me, I read the bileblog. Err, anyway. My point was that this guy is a snake-oil salesman that wants to use the linux kernel as his bathroom to dump his pile of excrement called reiserfs4 into the kernel.

Reiserfs4 won't be merged as long as it's doing everything differently from the other filesystems. If it's too different, it will fragment the APIs and that way lies madness that will take many a painful year to fix, and produce no end of complaints for Linux's clueless users. So the kernel devs are unrelenting on the point that as long as reiserfs4 does weird shit with Unix semantics and implements plugins on the wrong layer of abstraction, it won't get in.

But there's a chorus of clueless users screaming that reiserfs4 must go in now. They actually believe that Reiserfs4 represents the future of where filesystems want to go. And puts Linux on the cutting edge of everything. Hey, maybe it does. (I mean, err, clearly we should do new things with filesystems, but I think the real answer is in FUSE type things anyway. Userland is just so much more suitable for doing strange things than the kernel.)

Fast forward a few years. If there are merits in reiserfs4's approach, or some similar one, then all the other filesystems will move forward in that direction. However, if reiserfs4 starts now and does everything differently, it will be hard to fix to fall in line with the others, and this means greater overall system complexity! And in that way it will again make Linux more confusing for distributors, programmers and users. Everyone loses.

An anonymous person on kerneltrap wrote it perfectly in response to my comment that appears under "Anonymous Cow" handle:

As far as coding style and design go, the Linux kernel is basically
an advertising platform as far as Hans is concerned. It is the place
where his filesystems can go that guarantees high visibility and
usage. If you're a very ambitious filesystem designer (you might
even say "fame seeking"), what more could you ask for? It's obvious
that he cares very little for the design of the kernel as a whole,
provided it provides the necessary facilities for his filesystems to
work.

In other words, like this person, I believe Hans isn't in here to play nice with others. He wants the advertisement and the gratification for being able to think "I made the best linux filesystem in existence". And he seems to have some sort of messiah complex, thinking that the divine excrement he laboriously extracted out of his bowels is being unfairly treated, rather than being exposed for the shit it is. Witness his starting remarks on the recent thread on linux-kernel:

The VFS layering issue was addressed after 2 months of recoding.

The undesired type safe lists were removed after ~ a man week of coding.

What he wants to say is "we are here suffering losses of time and money due to arbitrary constraints imposed by others" rather than "our bad design cost us 2 months to fix".

Mr. Reiser is really a rather offensive person. I don't know why so many other people don't seem to realize it. Down with Reiser. Think rather how to improve the VFS component so that all filesystems can benefit. It's the only acceptable solution, anyway.