What follows is a letter from Hans Reiser to myself, which he wrote some two months back, and has asked me to publish, with his thoughts on the deprecation of ReiserFS from the Linux kernel. I have transcribed it to the best of my ability. Plaintext email may not be the best way to read it, as such, I have also made available PDF and HTML versions of the letter.
↫ Fredrick R. Brennan
Hans Reiser is the creator of the ReiserFS file system, which used to be a serious contender for the Linux file system you’d use in the early 2000s. In 2006, Hans Reiser murdered his wife, and is currently serving a prison sentence for this crime. Hopefully, after he completes his prison sentence, he can become a contributing member of society once again, if the professionals and specialists involved in such matters deem him capable of doing so.
The long letter mentioned here was actually quite a fascinating read, and details his abrasive behaviour in the Linux world, the design of ReiserFS and its place in the ecosystem at the time, and his thoughts on the removal of ReiserFS from the Linux kernel.
I used ReiserFS ~2003-2006 as my filesystem of choice, maybe a bit past then too.
It was a real shame he got stabby.
I used it for a few years. Thing was every time I did the seagate drive would fail within 2 to several months. My 5 year warrenty got a lot of use. I stopped having drive failures when I stopped using this filesystem. It was a neat filesystem yet clearly was hard on the hardware.
xeoron,
Do you have any direct or indirect evidence for this? Never say never, but it sounds like maybe a coincidence?
I had a batch of bad seagate drives too, every single disk in the raid array. They all needed to be returned for warranty in a short period of time. There was a firmware glitch and all the drives were failing if not updated. You may have encountered something similar?
“if the professionals and specialists involved in such matters deem him capable of doing so
This presumes that the job he takes up with presumably be a technical one… but pray tell who wants to work with a murderer. He can contribute to society in various other ways but to presume that WE technical people must accept a wife murderer as a coworker is frankly rather a bit of a stretch don’t you think?
I also used ReiserFS… it was fine as an FS not without flaws but it had some good points certainly as a desktop workstation FS. If anything I would have liked to have seen his code renamed and supported, after all IT didn’t commit murder. But the time for that has long since passed, there are other FS leading the pack now.
Working with a murderer in person is significantly different to dealing with a git commit over the internet.
He could use a pseudonym, no one would ever know but I suspect he enjoys the notoriety which was no doubt of the the appeal to him in his pre-murder endeavours.
Probably people in Texas, SV, 2A fanatics, or right-wingers anywhere.
Tech people accept jobs at ad-tech companies all the time. Or work in the defense industry. Tech is probably not the industry that wants to start casting moral judgements on people who have done their time.
The dude is a famous murderer… even if they hire him people may even quit to not be around the dude.
That was a very insightful read, almost harrowing at points.
These things always come round to “can you seperate art from artist”.
In tech (and especially FOSS) you often cant, as many many projects revolve around the contributions/leadership of a single person. Just look at the most famous example Linus/Linux!
The letter is worth a read and is a good reflection on mistakes young men do lacking patience and experience. Could I have sold my many ideas much better, when I was young? Hell yeah!
There are two more sad lessons:
1) there is a small thing line between genius and madness
2) never underestimate how important your wife is for keeping you on the right side of that equation (pun intended)
In quite a few countries, it is legal to work from prison. While most people work simple jobs due to limited skills, somebody could technically work as an open-source contributor.
It’s the US. It’s a punitive system.
Should we discount someone’s work just because they may have done horrific things in the past?
The nazis made significant contributions to science and medicine, including many horrific experiments but the data learned from those experiments was not thrown away. Many nazi scientists were recruited after the war and put to work.
There is also the notion of rehabilitation for prisoners. If someone is in jail, they are supposed to be paying a debt to society – what better way than to produce open source code if the inmate is able to?
This is not a black and white thing unfortunately. Let me explain:
a) does everyone deserve a second chance? Yes, absolutely and I actually read a lot of Goodwill into this letter.
b) do I know and understand, what really happened that day and all the emotions and circumstance? Definitely not.
c) however, real life projects are foremost about Trust, which is a very intangible and fragile good. Naturally I will trust my data better to a person with a flawless record of credibility.
That said, I am all in for “re-socialization” and “opportunity to redeem himself” over “revenge”. Unfortunately US does not seem to care much about those thoughts.
“the nazis …”
Very slippery slope, please be careful walking down this path even when with good intention. I have heard this “not everything was bad …” way to often from various cronies of the regime on both sides of the spectrum left and right.
There was NOTHING ever good about the Nazis or about the Bolschewiks, Stalin or Pol Pot or Mao or any other person coming down on minorities and individuals for making up their own failures, short comings and miseries.
True. The rocket that took the man to the moon was developed by a nazi scientist.
I remember that Reiser4 wasn’t incorporated into the kernel (among other things) because it’s plug-in system was not general for al the filesystems.
Also someone is working on Reiser5
https://arstechnica.com/gadgets/2023/08/the-torrid-saga-of-reiserfs-nears-its-end-with-obsolete-label-in-linux-kernel/