As a programmer ...

First I could not condone giving him access to the facilities necessary to 
carry on his work.  A convicted murderer shouldn't be given any sort of special treatment
regardless of his or her skills. (And giving a hacker access to a computer and Internet
connectivity --- and allowing him the freedom to work "hacker's hours" on such things would
count as "special treatment"). 

Frankly, I don't know if he did it or not.  But he has been convicted and I didn't see
anything egregiously unfair in the handling of his case. (The lack of a corpse and the
estranged victim's foreign ties might have been "the shadow of a doubt for me" but I wasn't in
the court room and didn't see the evidence; and I would have been recused from the jury if I'd
been called up for it --- since I was an aquaintance).

Regarding possible impact to ReiserFS and more generally to Linux ...

Hans was the founder, and principal sponsor of the project.  The design concept was his and it
embodies an innovative unifying idea.  I admire the concept.  He was not the main programmer.
From what he said, on various occasions when I heard him speak, both formally and around the
dinner table as SVLUG after-meeting social events, he hired programmers, mostly Russian, to do
the bulk of the implementation.  I don't know if he did any of the low level coding personally
--- but his own description of the project to me suggests that this would have been relatively
minor.

This isn't intended to disparage his work.  It's just the facts as I understand them.

I'm a bit disappointed that ReiserFS development and adoption has lagged so much.  However, I
can't say I'm surprised.  It has suffered some horrible technical setbacks (quite apart from
the sordid issues related to its namesake and some of the vitriolic verging on childish flame
wars).  (The debacle of a filesystem whose fsck utility would merge looped filesystems out of
their containment files and into their parents would be just about the worst that I've heard
of in any production Linux code).

I hope that development will continue ... and that adoption will find its niche --- or that
some of the design concepts (particularly the tail-packing and ubiquitous indexing) will be
merged into other filesystem designs (ext5 or btrfs++ or whatever).  But, if that's to happen
it will have to be without Hans' involvement (at least for the next 25 years or so).

JimD