ReiserFS stuck around for almost two decades after his crime. The reason it’s being removed has very little to do with his crime and more to do with the stability and maintenance capabilities of it.
For reference, his wife was missing in 2006. He was prosecuted in 2008. It was deprecated in 2022 and scheduled to be removed in 2025.
How does someone believe that it’s due to his history when there’s been almost twenty years of grace?
The reason for removal has been very clear by the Linux maintainers.
There was a consensus among upstream kernel developers to deprecate ReiserFS given that there are no notable users of it any longer, the code is barely maintained, and no legitimate reasons have been found to user ReiserFS over any of the modern alternatives like EXT4, XFS, Btrfs, F2FS, etc.
That's not entirely fair, being included in the trunk of linux for a long time is normal even if something is unused... I know people were avoiding ReiserFS because of its indeterminate future in the backtrack linux 2.0 days (March 2007).
It's a cart and horse problem, nobody wants to use it because they think there's no investment, since developing it could be seen outside as being an indemnification of the person.
Lack of use means that less people even want to invest in developing it further, making it take a back seat, which then leads to even fewer scenarios where it's used, because it's not competitive, and so on.
The Linux kernel rarely removes stuff. This is not something that should be taken as a measure of support; If we want to track reiserFS properly we must follow the commits before prosecution and after, along with some usage trends would paint a real picture.
There's a whole host of dead things in Linux (AFS, which even Apple doesn't support for nearly 2 decades) that isn't removed yet.
I can’t find a reference to a filesystem called AFS. Would you have a link?
As far as I know there was Macintosh File System (MFS) which was replaced by HFS/+ and finally APFS.
But ReiserFS was also infamous for its difficult to maintain nature at the time of its inclusion. Linus himself was against a lot of aspects of it. They capitulated to community pressure to include it, but it was always going to be a project with a high bus factor regardless of his crime.
AFS, the Andrew File System, has kernel support. [0]
AFP, the Apple Filing Protocol, is a network protocol, not a filesystem. It has several userspace implementations. [1]
AFS, the Apple File Service, is AFP plus other auxiliary protocols. [1]
APFS, the Apple File System used in modern MacOS, has read only support in userspace, and there is work on a read/write kernel module that is not mainlined. [2] [4]
HFS, the Hierarchical File System used in older MacOS, has mainline kernel support. [3]
Ah the SMB alternative. Okay yes I’m familiar with that.
Unfortunately it was still frequently in use 10 years ago on Apple systems (particularly the time capsule which was only discontinued in 2018) , and there are derivatives in use today. So I’m not sure it’s an apt comparison point.
The Wikipedia article actually links to implementations like netatalk that have updates as recently as a couple months ago. So it is actively maintained beyond what ReiserFS has.
netatalk and AFS are similar but truly not the same, netatalk in linux does not afaik have an actual kernel driver.
AFS however, does, and I was doing MacOS sysadmin work in 2011 and it was strongly discouraged to use anything other than SMB, which frustrated me because samba3 was a pain and I considered SMB a windows technology at the time.
The time capsule as you mention also discontinued the usage of AFS after the first version (I actually have 3 at home right now still, a 1st gen, 2nd gen and last gen - the tall one).
AFS's use today, not sure, but as mentioned, 13 years ago at least it was strongly discourage by Apple themselves.
AFP was still the default though in 2012. And even though you weren’t able to provision new time capsules to communicate over AFP, it was still supported.
I’d bet there’s a significantly larger set of devices active at any given time during the existence of ReiserFS to today that still use AFP.
Well, the "simple, incomplete" Andrew Filesystem is actually an even better example then, as it cannot possibly have users as it's incomplete but has been in Linux since 2004.
Fair enough, though I think that even if you don’t explicitly say it, the whole Artist vs Art framing only leaves that interpretation. As can be seen by the rest of the responses here.
I think cmiller below nailed what I was trying to draw attention to:
My takeaway was that Han's problems were intrinsically tied to the both the architecture and code in ways that made it difficult for others to contribute and fix. The artist's influence on the art caused its ultimate demise.
So is about art and artist, but no causation implied with Linux depreciation and the murder, but I get how people might think that.
But also I just wanted to share the story with those who may not have heard it.
i still use reiserfs - i can store more tiny files than the others. I was disappointed when testing Btrfs as it used more space and became corrupted easier, even when using compression.
ReiserFS is named after the author, which is an oddity among file systems or open source projects in general. I wonder if it would have seen further development if it had been named something else.
I don't know how much of an oddity it is, but Linux is probably the most known FLOSS. OK Linus himself wanted to name it "freak" or something like that.
Meanwhile we have a graphics suite called gimp, a 3D model loading library called assimp and if someone calls for blender, you can't be sure if it's the kitchen utensil or the 3d creation suite. :)
and a version control called Git. The old tech world was many things, but "politically correct" is far down that list.
I'll be fair and note that Assimp pre-dated the modern vernacular of the obvious word that comes to mind and was actually reasonable (Open ASSet IMPort Libray).
The Linux reboot system call supposedly accept Linus' daughters' birth date as a magic value [0]. It's not like that prevented people from rebooting their systems.
Maybe I am too numb, but I do not see problem here.
People in prison should be allowed to work. He is serving his sentence. Where is the problem?
One day he will be out of prison. Does it mean he should never ever be allowed to work again?
US has a problem with "prison vengeance", where prisoners are not allowed to even have a TV. But it would be cheaper and decrease overall crime. Current prison system is just raising more criminals.
>US has a problem with "prison vengeance", where prisoners are not allowed to even have a TV. But it would be cheaper and decrease overall crime. Current prison system is just raising more criminals.
we americcans don't rehabilitate. We just want homeless off the street and out of mind. Clearly if you did a bad thing in life you are a bad person and deserve everything taken from you. California (a more progressive state) had a propostion this election cycle to simply, as I'll quote from a summary:
>This proposed amendment to California's constitution would bar slavery in any form and repeal a current provision allowing involuntary servitude as a punishment for crime.
and it didn't pass. This is already federally illegal, but even the idea of taking some time to clarify the language wasn't popular.
Meanwhile prop 36
>This ballot measure would increase penalties for repeated theft offenses and certain drug crimes, including some involving fentanyl. It also would create a drug court treatment program for people with multiple drug possession convictions.
Passed overwhelmingly with 70%. Nevermind the government is dismantling those who spread the Fentalnyl into hospitals with recknless abandon as we speak. Just gotta get them druggies off the street. Their fault for getting injuected by doctors, I guess.
-----
I guess as people say these days: "It's not about winning, it's about sending a message". If even half the amount California spent on "hard on crime" simply went towards getting these homeless a job, the problem would essentially be solved for many. Then half the remaining can go towards rehabilitating those with genuine addictions and other mental health issues.
I've read that the American prison system is punitive instead of rehabilitative. The punitive system has a higher reincidence level, which I guess is good if you want to keep a specific portion as indentured servants.
I'm very glad there is a transcript so I didn't have to listen to it.
It seems like in this case ReiserFS is being deprecated and removed because it is no longer used and not because of who the creator is? https://www.phoronix.com/news/ReiserFS-Deprecate-Remove-2025
If so, then this isn't really about separating the art from the artist?
My takeaway was that Han's problems were intrinsically tied to the both the architecture and code in ways that made it difficult for others to contribute and fix. The artist's influence on the art caused its ultimate demise.
It’s not just Hans. Who hasn’t known some leading artist who was like this? Or even some former colleague? I’m sure most of us “artsy” types go through some phase like this, but this guy made it his whole life, and never questioned it until it was too late, too little. The art and the artist thing is real.
H. P. Lovecraft is a good example from literature. He was profoundly racist, even beyond the cultural norms of his time. However, as he has long since passed away buying and amplifying his works do not further his views and causes.
A modern counterexample would be J. K. Rowling. Where supporting her works and properties does directly contribute to furthering her prejudices in a very real way.
buying and amplifying his works do not further his views and causes
i'd rephrase that. it does not support the artist to allow him to continue spreading his views and causes, but it still draws attention to him, and potentially lets people learn about it. even this discussion here. i would not have known about lovecrafts or rowlings views if it weren't for reading about it on hackernews some time ago.
however an even more important concern is how much of these views are woven into the stories. as such it is important to at least be aware. i have ignored harry potter until now but my kids are getting interested, and so i am keeping a close eye on what they are watching for that reason.
First thing is to differentiate with how long the person is dead.
For example, when I accompanied my daughter class to a museum last year, there was a sewed reproduction of the Minotaur by Picasso (the artist exposed was a woman and this was just a collaboration she made with him). To say the least, the history behind the work is not the most glamorous. Would you explain the context to a 7 years class? But Picasso is so close in time, that his direct descendent have financial interest in exploiting the artistic legacy.
Now if you consider some artist like those who made graphic arts in Lasceau cave, of even someone as close and individually nameable as Katsushika Hokusai that died before the world wide madness of "intellectual property", that's a very different matter.
> But Picasso is so close in time, that his direct descendent have financial interest in exploiting the artistic legacy.
I wouldn't be concerned by this specifically. His relatives likely suffered from him being a jerk, I don't mind them at least benefiting from his work. I'm more concerned about Picasso being put on a pedestal.
I don't think I would be against exposing his art with the whole context though.
There can still be a moral concern if the artist is seen as a (role) model / genius and consuming / promoting their art causes the artist to be seen as a model for longer, potentially making it look like what they did is okay or forgiven given the art. We totally need a strong signal that doing good art doesn't forgive or allow being a jerk so jerks are not encouraged to take this path.
Another thing to have in mind: beside moral concerns, often, you can't separate the artist form the art because the art reflects the artist; you'd miss out on the interpretation of the art.
Sometimes the artist put too much of themselves in the work. For example: I tried reading Orson Scott Card's Iron Man comics and there was just too much homophobic nonsense throughout.
i stopped reading the ringworld series by larry niven when i found that he kept repeating how sex was used to seal a business transaction. it added nothing to the story and just seemed like wishful thinking from the author.
i can't blame them for it. it's only natural. when i write there is a lot of my personality in it too. in part that is the point and in many cases it is what makes a work worth reading. unless it makes the story unreadable like in our examples.
I read the whole series and don't remember this at all, even a single example let alone appearing so repeatedly that it became noticeable and annoying.
I'm not saying I don't believe you. Whatever it is you're talking about is probably in there.
I'm saying maybe you were just super sensitive to something that was actually insignificant.
that's possibly true. it wasn't in all of the books, but at least in the one where i stopped reading. and it was insignificant because it served no purpose in the plot. what bothered me was not the reference to sex but the fact that it made no sense in the story and that it was treated as something as casual as a handshake and it simply felt like it was the authors personal fantasy.
and if i may say so, i suspect that in general reference to sex is so common that many of us don't notice. it doesn't bother me but i simply don't care for it when it's not a significant plot element. i grew up without any exposure to this kind of theme.
I agree. We all make our own decisions. I won't say someone is wrong for making a different decision. For me, there are people who's product I won't use because they are a jerk and I don't want to contribute to their financial success. But if that same jerk was one of the creators of a commonly used language that is ubiquitous I won't avoid it because that person doesn't financially benefit from it.
If a murderer creates a file system/energy source/mouse trap that is effectively better than all of the alternatives, and generates value to the world, what he did in his personal life is effectively moot. There is no ethics problem involved in choosing the best solution to a problem, when that solution exists.
If financially supporting said murderer allows them to continue to possibly murder more, then I would say there is a moral dilemma. Less so when the murderer is in jail and unlikely to ever really benefit from your using said solution.
Which gladly rarely happens, so often you still have sufficient choice. Now, if it turns out Euler sabotaged some bridges and caused deaths, you'd still have a hard time avoiding using e.
But for software? It's usually rather easy to find alternatives. And ReiserFS never was an exception here.
The idea of “stochastic death” is a dangerous and unhelpful escalation of rhetoric that amounts to an excuse to call someone a murderer because they said something you disagree with. It’s the idea that Biden was somehow a “partial murderer” because he said something that might have, possibly, maybe, influenced a wacko to attempt an assassination on Trump. If we start calling everyone a type of murderer, it cheapens the word, dishonors the victims of actual murder, and hardens hearts with hate and fear of the dreaded Other when what the tribalists need most is a little mutual understanding.
Bill Gates multi decade marriage came to an end after his closeness with Epstein became public. I don't think anyone stopped buying ms software because of it.
Ok, he no longer works there, but I'm sure he still benefits from it.
If a video were released of Bill with a minor then I'd expect Microsoft to take a huge hit. Though it'd have to be irrefutable video. Because we've seen other Epstein accomplices thrive despite circumstancal evidence, like a certain real estate mogul.
Reiser4 is still in development and it is the successor to ReiserFS. It's probably a good idea to rename the filesystem in order to remove as much as possible any historical ties to Hans Reiser. It's time to move on and forget about Hans.
I was always a fan of reiser3, never had any knowledge or opinion of Hans.
But listening to what he has to say towards the end here, even now after all this time, all I hear is:
He expresses remourse that he wasn't smart enough to press the right buttons to get the right responses out of everyone else.
He regrets that this miscalculation resulted in him failing to achieve his goals.
He litterally only says it that way. Not "It must have really sucked dealing with me. What a dick I was...", just "it was a mistake not taking other people's feelings into account"
The kernel credits are interesting.
That is a little more possibly genuine. It seems genuine, that he really wants them to be acknowledged simply because they should be and not because he calculated that it makes him look better.
But I don't know how materially valuable that might be for parole reasons. If it might make even a little difference there, then you can't trust it, because this is not a benefit of the doubt situation, this is a known entity who has already manipulated others many times, and so now it no longer matters that an act might finally be genuine this time, it makes no sense to assume that in this case.
But in my own example of pressing buttons, I try not to shit on behavior I want to see. All else being equal I want to just accept that small act as good on it's face and leave it at that.
Also because it seems petty not to, because he gets nothing else at least from me, forever, because his wife can't be forgiven and redemptioned back to life.
I didn't know about this aspect of the filesystem. Seems like WinFS executed better than Windows Longhorn/Vista.
> Hans: Reiser4 is not only a file system. It is a software framework for creation, assembly, and customizing file systems.
> It’s a bit hard for me to understand his vision. But I think its a bit like replaces a file based system with a database, where files can easily be searched and indexed and have metadata and plugins can add whole new ways of seeing the file system layer. It’s a lot.
Can or want? I don't think it's a question of capability, but a question of conviction. At least for some of us.
Separating or not can have real consequences. The question is what consequences do we prefer?
You are suggesting the status quo. You can do this for any topic, but I don't believe this helps progressing much (of course, you might be uninterested in this, but I could still be interested to show you that you might want to reconsider). If you believe there are concrete consequences, you may want to convince people your way, and this seems valid too.
I agree with not forcing. You can't anyway. But arguing and debating ideas still has value. It can make people change their views, or at least be interesting.
You can't reject someone's idea that the art can be harmful because of the artist as personal preferences or way of thinkingnfor instance. The more interesting thing to do will be to prove them wrong.
Back in the day on Slashdot the Reiser/ReiserFS saga was one of the more interesting linux-related dramas. Seems like the other FS options in the kernel are now unequivocally superior and ReiserFS is completely obsolete.
Hey, I made this. It uses OpenAI voices for Hans and others, which some found too robotic.
The topic on my mind around this was: in the large, like MJ or Woody Allen, it's hard to make a call on the art vs. artist.
But in the small, it's much easier because people aren't divided into domains. It's all mixed together. And if you can't work with people, if you can't be a team player you limit your abilities.
ReiserFS4 didn't get merged because of his empathy issues, and those issues are also presumably contributors to his murdering his wife.
So it's all wrapped up together. Making stuff is a team sport. Even people like Fabrice Bellard seem to be successful because of collaboration. Everything uses FFmpeg, but he's no longer the main contributor. It's successful because of people working together.
> The topic on my mind around this was: in the large, like MJ or Woody Allen, it's hard to make a call on the art vs. artist
Why do people think this is a dichotomy, or a call to be made in the first place? Instead of artist being an inseparable part of the art itself?
Chuck Tingle put it pretty well:
> strange when buds talk on 'separate the art from the artist'. a painting is not just a square that ends at canvas edge. it is history and intention. if you build wall between them you are numbing yourself and only experiencing half the art. the artist IS THE ART, at least to me
> separating artist from art is like walking out of a play during intermission youre missing half the show bud, for better or worse. gotta accept theres a whole second half and grapple with that because its all one big performance
Detail, but I think this part here is slightly off:
> Fast forward to March 1998. Saint Petersburg. A cafe next to a canal. Hans is meeting a woman. Nina Sharanova. A mail-order bride.
According to his Wikipedia entry, Nina was the interpreter involved when he arranged to meet a mail-order bride:
> In 1998, while working in Saint Petersburg, Russia, Reiser reportedly arranged to meet a Russian woman he selected from a mail-order bride catalog.[15] Five months later, he married the pair's interpreter, Nina Sharanova (Russian: Нина Шаранова).[16]
Doctors listing themselves in mail-order bride catalogues seems at least as unlikely to me, but admittedly I know nothing about the usual demographics of this phenomenon
Some part of me always assumes people are innocent. I think I just see too much of myself in others and when they commit some act I never would it seems genuinely outside reality.
I remember vividly being a very tech-involved twenty year old, extremely excited about Reiser4 when this all came to light and thinking "There's no way he did it! They don't even have a body, they don't even know she is dead" right up until he lead the investigators to the body. Then I felt so dumb.
The whole thing honestly lead to a lot of self reflection and questioning of my own instincts. Frankly, it made me a little more cynical about people overall. That said, I am still a perennial optimist.
> Some dumb part of me that always thinks people are innocent.
You're not alone. Hans Reiser is one of two people where I had the notion that "Why would he do that? That makes no sense." The other one is Peter Madsen (the Danish submarine guy), he had a few month earlier been part of a TV show about find some German WWII u-boat, where he was absolutely brilliant and entertaining.
For ReiserFS a large part of the issue, which is also addressed here, is that there just wasn't that many developers that cared that much for the filesystem. The Namesys developered cared, in large part because they where paid. Obviously they also had bills to pay, so when no one wanted to pay them to work on ReiserFS it just feel apart. We see this with many of the project that thrown into e.g. The Apache Foundation, the code is thrown over the fence, but with no advocates, no developer resources and no money, so the code is just left to rot.
Narcissists can be very charming. You don't know the inner lives of most people and it's impossible to judge them by their public persona when it's entirely under their control. Celebrity actors are notable specialists in this area.
> where he was absolutely brilliant and entertaining.
When I was young and naive I thought con artists would be easy to spot: They would be obviously bad people and they would lie in everything they said.
The challenge with successful con artists is that they tend to be very charismatic. They might also tell the truth 99% of the time. Telling mostly truths is the key to building trust, after which they can slip in a lie here and there and it will go unquestioned.
This is a major problem in pop-culture science and health discourse right now. There are a lot of podcasters and bloggers who spend most of their time repeating simple and undeniable truths (exercise is good, you should eat more vegetables, getting sleep makes you feel better, going outside is a good idea) and therefore build up a lot of trust. Then they'll change gears to wacky alternative medicine ideas and slip them in between all of the obvious truths. For people who consume hours of their podcasts every week, those alternative medicine and pseudoscience claims are hard to identify among all the other simple truths.
> I remember vividly being a very tech-involved twenty year old, extremely excited about Reiser4 when this all came to light and thinking "There's no way he did it!
Becoming involved with a person and/or their project has a way of clouding judgment. ReiserFS blurred the lines between person and technology, so getting excited about the file system in a way meant you were backing the person, too.
None of us likes to imagine ourselves supporting a murderer, criminal, or other bad person. So when the accusations come out, the first instinct is to deny them. Admitting they have any truth would be equivalent to admitting that we supported a bad person, but denying them allows us to remain consistent.
This plays out frequently with celebrities, internet personalities, and politicians.
More recently (and without the crime, obviously) I'm reminded of the revelation that Andrew Huberman was not a great person in his personal or professional lives, and that his podcast content is full of misleading claims. The poor accuracy of his scientific content was well-known in the research community, but people didn't want to believe it because they liked Huberman and his personality. When the revelations about his personal and professional activities came out, people rushed to his defense and dismissal. Even today mentioning his name online will summon commenters who downplay the accusations or try to defend him. Listening to hundreds of hours of someone speaking charismatically has a way of charming people into denying any information that goes contrary to what they want to believe about that person.
It seems there were cracks earlier in the software story:
> Linus: As long as you call them ‘plugins’…I (and I suspect a lot of other people) are totally uninterested…
> Uninterested, because they’re worried about the complexity, the stability, the long-term viability of Reiser4. They’re worried about his inability to collaborate, to compromise. They’re seeing a man who’s so focused on his own vision that he’s blind to the practical realities of working within a community, of releasing and maintain code at the scale of heavily used operating system kernel.
> Alan-Cox: “It doesn’t matter if reiser4 causes crashes. It matters that people can fix them…and the code is maintainable.”
When I read about politics I have to accept that almost half of people either have a completely thought process from me, or mediated communication allows them to live in a completely different world
ReiserFS stuck around for almost two decades after his crime. The reason it’s being removed has very little to do with his crime and more to do with the stability and maintenance capabilities of it.
For reference, his wife was missing in 2006. He was prosecuted in 2008. It was deprecated in 2022 and scheduled to be removed in 2025.
How does someone believe that it’s due to his history when there’s been almost twenty years of grace?
The reason for removal has been very clear by the Linux maintainers.
There was a consensus among upstream kernel developers to deprecate ReiserFS given that there are no notable users of it any longer, the code is barely maintained, and no legitimate reasons have been found to user ReiserFS over any of the modern alternatives like EXT4, XFS, Btrfs, F2FS, etc.