Right? Imagine running a company under the name Hitler Inc. after the end of world war 2. The name just needlessly taints the project which is (obviously) not run by the convicted murderer anymore.
I agree. I used to use reiserfs on everything I touched back in the day: Mandrake, SuSE etc. When Hans R became toxic and ext4 happened to suddenly fit the bill, I followed the rest of the herd and reiserfs3 rapidly fell off the top n filesystem charts. I remember reiser doing some things much, much faster than say ext2, eg directory listing with lots of files. It also seemed to know when to repair itself on boot and simply do the right thing - that would be the journalling that ext2 didn't have and ext3/4 took a while to get working properly.
They could call it rfs or something, anything but reiserfs.
Aside:
"Quick" can mean "alive" - I think that was its original meaning. We should also note that schnell is often translated as "quick" but I think "faster" is better in general. Using a simple dictionary approach to translate mordsschnell into English might come unstuck pretty quickly. With these clues we could arrive at living dead or zombie! I don't know but given how fast the above comment got downvoted that word has some history.
It wouldn't be uncalled for. The past ten years' worth of patches have pretty much come from Edward Shishkin. I can only imagine the work required to keep such a large codebase maintained out of kernel during that time.
The ideas described in the linked email looks quite original, too. The abstraction level between blocks and files have always been a leaky one. It would been interesting to see them implemented. I guess it would require a lot more RAM to be effective but that might not be an impediment these days.
Why? That’s just a name. If people are that reluctant to use a technology just because a murderer used to develop some parts of it, what would they think of current developers who renamed the project just to conceal that historical fact? Wouldn’t they initiate a grand purge once they learn that somethingelsefs is actually a project that uses code and ideas of reiserfs?
I don't see the relation between the country where they live and them changing their names. I couldn't find any news to that effect either. Seems like a bit of a stretch.
So... how changing a name averts that situation? By hoping that the kid never finds out? And should current developers rather abstain from continuing development as they obviously behave as insensitive egoistic jerks that instead should drink some brain bleach, bury the project, and move on? Come one, that’s ridiculous. It’s more irrational than blaming the gun for the murder: it’s like blaming the word for that and pretending that if you call it something else then it did not happen.
you're still not making the attempt; it requires imagining being that person. my point is very simple and obvious: having to constantly be presented with the name of your abuser is painful.
the implication is that even though i haven't had my mother murdered i can simulate the experience and thereby relate (to an extent) to someone that has and make decisions with those insights in mind.
Edit: I just realized that what you probably meant is that if it's never happened to me then I can't genuinely be concerned. I didn't even know that that was something people believed. Interesting. Why would anyone pretend to be concerned for someone else's wellbeing?
Ah... "virtue signaling". What the sociopaths among us have branded empathy they cannot feel or understand in others. Just because it is not an emotion you have, does not mean it is not real for others.
>Imagine we spent all of our energy avoiding already unlikely situations that might marginally harm ~2 out of 7.5 billion people on earth.
false dichotomy. this isn't a choice between not doing it and doing it absolutely always. i'm not claiming we need to spend all of our energy.
i think today we spend very little of our energy optimizing for any kind of empathy let alone marginal. so any increase is still well within manageable (in terms of cost).
That's a rather hypothetical situation, his kids may as well avoid programming altogether or simply be interested in something else.
Your comment, by the way, is very insensitive, surely with your empathy you could've imagined possibility of the author of the parent comment being autistic and hurt by your suggestion that empathy is easy.
In my experience, autistics are very empathic. Sometimes in different ways than neurotypicals, and how it's displayed might be different as well.
Googling something like "quora autistic empathy" will give you plenty of informed posts about the subject. I recommend reading the ones written by actual autistics.
You can try rationalizing, but the name immediately raises very unfavorable emotions and associations.
The very first thing I thought about when I saw the headline was Hans Reiser's crime. The first thing I then expected in the comments was a lengthy discussion about that. I am fairly certain that this is fully in line with how most people react.
I don't see how your proposed hypothetical following a name change would take hold. I don't think a name change would be perceived as "concealing", just understandably moving away from a name that has become unfortunate. I think people are more surprised that they haven't.
Branding isn't a very logical science; so sure they probably should. Them and the GNU Image Manipulation Program, which has 100% lost users because of the stupid name.
That said, we also need to be able to separate the good and bad people do. Reiser in his capacity as a programmer is really a different entity to him in his capacity as a murderer. Celebrating one is not celebrating the other.
It is the stereotypical example, but it is such a good one that it deserves to become a cliche. Fritz Haber was sufficiently evil that apparently his wife committed suicide because of him. The Haber–Bosch process is probably one of the greatest technical contributions to humanity, ever, full stop. Maybe the steam engine can compete or a handful of other things. We can't allow technical excellence to be shunned because the technician had horrible problems - the shunning lessens us all. We can afford for bad people to be remembered if the price they pay is to create something of great use.
GIMP changing its name seems pointless at this point, especially because of where the conversations there led. It would have been smarter for them to argue that "gimp" is a sexual term, and therefore gross, rather than an antiquated ableist term, because the former is more common than the latter usage (which is never used).
Reiser's accomplishments are already out there, there's no need to continue with his name on something he hasn't contributed a patch to in a decade, and there's literally no harm in changing the name of a program like this; it's not a binary decision between "remember the person" and "use their work," plenty of people have their work utilized despite no one knowing their name. I think that's fine.
> 'Reiser's accomplishments are already out there, there's no need to continue with his name on something he hasn't contributed a patch to in a decade'
He's been in prison. I doubt kernel driver writing is encouraged there.
> 'there's literally no harm in changing the name of a program like this; it's not a binary decision between "remember the person" and "use their work,"'
You are literally incorrect in saying there is literally no harm. The harm is this: you're engaging in the "Two Minutes Hate." I don't want to live in the kind of corrupt society that rushes out to scrub a person's name from the history books just because a skeleton is (literally) discovered in their closet.
Removing Hans Reiser's name from his creation would be equally immoral to putting him to death--because among other reasons, you cannot guarantee that he is guilty with 100% certainty.
I think that the name should stand, so that the community can collectively enjoy the constant whining and bitching from all corners every time the subject comes up, rather than anyone having a technical discussion about boring things like the filesystem's merits.
I'm not convinced by the long term benefit of nitrogen fertiliser synthesis. A significant boom in food production is attributable to it but also long term degradation of agricultural soils, pollution from fertiliser run off and repression of more sustainable agriculture and food distribution options. Perhaps if it had been used to buy time for more ecological and equitable policies it would be more impressive, but it seems only to have advanced the alternative.
You're right on the money. A major, growing world problem (that noone talks about, because they're too busy focusing on manufactured frauds like 'global warming') is depletion of soil micronutrients.
Zinc, selenium, boron depletion are some particularly serious examples. The vast majority of the world's agricultural soil is zinc deficient, including most of US soil.
This of course leads to deficiencies in our food also, with many foods in the USA being seriously depleted of nutrition, and the corresponding health effects in the populace. It's a slow-motion downward spiral, and existential crisis.
I understand other nutrients like iron are deficient in some of the more serious locales. Fortification of foods with various nutrients has been tried as a stopgap measure, but this problem is only going to worsen as time goes on.
The plain fact is, industrial agriculture is a dead end road, as you allude to in your comment about synthetic nitrogen fertilizer.
Phosphorus (rock phosphate) is an equally major problem, in that it's used way too much, tends to wash out of the soil and cause its own pollution, and is in decreasing world supply also.
Likewise, potassium is essential in large amounts for plant fruiting and flowering. There isn't really enough of it out there in good, usable, convenient forms.
The only answer is to use manure fertilizer, i.e. organics, and the end of industrial farming. Not going to happen without a major disaster occurring first, because the masses aren't known for their deep thinking, and the people running the show aren't known for anything other than total selfishness, and steering nations purposely into disaster for the sake of their own power or profit.
I have no idea how your post got here on a post about ReiserFS, unless it's an HN bug or something.
On the contrary. Once upon a time ReiserFS was a serious contender to be the next de facto standard Linux filesystem. Then Reiser did what he did and interest in his filesystem promptly became nonexistent, where it remains to this day.
That’s not marketing. It’s more like anti-marketing.
Reiser is the only filesystem on Linux that I ever lost data to. And they really should drop the name, that's not just a wish but a must, keeping it is completely tone-deaf. I was all for giving Reiser the benefit of the doubt as long as he wasn't convicted but it looked bad from the start and the way he lied about it means that that name can not today or forever be associated with something that requires trust in any form. Funny how that works. Lesson: do not name you OSS project after yourself, just in case one day you are convicted of murder.
it's sort of self-fulfilling-prophecy to talk about how bad the name tarnishes the project, while simultaneously explaining to everybody your opinion about why the name is bad.
anecdote : I asked a young developer near me 'Hey, do you know the Hans Reiser story?'. The reply was 'Who?'. The reply came from someone familiar with the filesystem, and that general line of work.
tl;dr : Reiser is just a name. It was recently the name of a 'bad guy', but it hasn't always been, and won't always be.
p.s. I've lost data with ReiserFS too. I'm glad I don't have any need for it any longer.
Caring about the name is neurotic and childish. When it's no longer a filesystem named after a murderer, it will still be a filesystem inspired and created by a murderer. And it'd be equally stupid to refuse to use it on these grounds alone. Those complaining are indulging in the thrill of cheap activism more than engaging their rational mind, and giving into their demands merely enables pointless noise-making at best and at worst conditions (the few-to-no) emotionally unstable people genuinely offended by the name to embrace their fragility at the expense of the project instead of reaching this healthy, antifragile state of mind I inhabit.
You are really insulting your fellow commenters here.
If you truly inhabit a healthy, antifragile state of mind, you must realize that many of us are not as enlightened and do have emotional reactions to things like this.
It doesn't mean we are inferior, it just means we are human.
This fragility does not make you a lesser person; I extend my hand to you as an equal, my brother. I hope you do the same when I am misguided or weakened in any manner myself.
The Reiser filesystem will continue to be used by those who understand its value, despite its bad reputation within a small niche of people[0].
On a more on-topic point: my experience has led me to believe that the Reiser fs (Reiser4) has a narrow use case only when quick operations (seek/read/write) on an inordinate amount (hundreds of thousands+) of small (<1MB) files is required -- and that high performance is absolutely necesaary.
I failed to find any other use in my experiences with it.
I was quite surprised when I initially ran out of inodes my first go-around, as well.
[0] Really, no one cares that much. Whenever the name comes up in less formal circles, obligatroy low-effort "dead wife" jokes surface, but what usually follows is discussion on the filesystem's merits. The outrage is from a vocal minority that I would deduce aren't in any position where choice of filesystem is of their concern.
The fact that every single thread here, possibly every single comment, is so far referring to the name of the filesystem, speaks louder to me than your anecdote. (It's very hard, for me at least, to say that in a way that does not come across as snarky, which I don't intend to be.)
I think you might just be trolling, but in case you are not, I do not see how selection bias applies here. I (or anyone else) did not select any comments here. Close, or maybe even actual, 100% of the comments here are about the issue with the name, so unless you want to insinuate that the HN community responding to this is not in any way representative of people who care about filesystems, there is hardly a bias here.
You imply that comments are representative of the attitude towards reiserfs in general population.
It seems more likely to me that overly emotional people, regardless of their interest in filesystems, got triggered and wrote a bunch of comments even without clicking the link.
How long do you want to wait? I don't particularly care about the name, but I was around at the time of the Reiser case, and I'd like to hope that I've got another 20-30 years in me yet.
However the reception of brand names is surely often self fulfilling like that, they don't succeed through robotic reason but through fuzzy associations. The murderers connection to the product is very direct so it seems it wont be forgotten soon. An unconnected product might use the name without being tainted.
I'm extremely neurotypical and I also don't care what things are named. People who get hung up on this are bikeshedding, commenting about murderFS for twenty years and holding back genuine innovations because of their thin skin and lack of actual inclination to talk about technical details.
The corollary is that naming is important in the adoption and proliferation of a technology. So public image is a concern technology devlopers should be on top of.
Don't hate the player, don't hate the game; Adapt and move on.
I think the approach on managing the block is quite inventive. I'm not sure if the way its handle atomicity is safe enough, will be very interesting to get some benchmark on popular database.
87 comments
[ 3.3 ms ] story [ 162 ms ] threadhttp://www.iaauae.org/media-files/2007/10/16/20071016_Hitler...
They could call it rfs or something, anything but reiserfs.
Mordsschnell!
Aside: "Quick" can mean "alive" - I think that was its original meaning. We should also note that schnell is often translated as "quick" but I think "faster" is better in general. Using a simple dictionary approach to translate mordsschnell into English might come unstuck pretty quickly. With these clues we could arrive at living dead or zombie! I don't know but given how fast the above comment got downvoted that word has some history.
The ideas described in the linked email looks quite original, too. The abstraction level between blocks and files have always been a leaky one. It would been interesting to see them implemented. I guess it would require a lot more RAM to be effective but that might not be an impediment these days.
Not a great namesake, is it?
is it really that difficult to attempt empathy?
imagine being one of his kids, growing up to become a developer, and then having to work on code that intersects with reiserfs.
since kids are in Russia i highly doubt that.
https://www.merriam-webster.com/words-at-play/sympathy-empat...
the implication is that even though i haven't had my mother murdered i can simulate the experience and thereby relate (to an extent) to someone that has and make decisions with those insights in mind.
Edit: I just realized that what you probably meant is that if it's never happened to me then I can't genuinely be concerned. I didn't even know that that was something people believed. Interesting. Why would anyone pretend to be concerned for someone else's wellbeing?
Virtue signalling.
Imagine we spent all of our energy avoiding already unlikely situations that might marginally harm ~2 out of 7.5 billion people on earth.
And what about the kids whose parents were murdered by people with more common names?
false dichotomy. this isn't a choice between not doing it and doing it absolutely always. i'm not claiming we need to spend all of our energy.
i think today we spend very little of our energy optimizing for any kind of empathy let alone marginal. so any increase is still well within manageable (in terms of cost).
Your comment, by the way, is very insensitive, surely with your empathy you could've imagined possibility of the author of the parent comment being autistic and hurt by your suggestion that empathy is easy.
Googling something like "quora autistic empathy" will give you plenty of informed posts about the subject. I recommend reading the ones written by actual autistics.
Here are a few examples:
- https://www.quora.com/How-do-you-think-autistic-people-deal-...
- https://www.quora.com/Can-autistic-people-be-empathetic
- https://www.quora.com/Can-people-with-ASD-Asperger-Syndrome-...
The very first thing I thought about when I saw the headline was Hans Reiser's crime. The first thing I then expected in the comments was a lengthy discussion about that. I am fairly certain that this is fully in line with how most people react.
I don't see how your proposed hypothetical following a name change would take hold. I don't think a name change would be perceived as "concealing", just understandably moving away from a name that has become unfortunate. I think people are more surprised that they haven't.
That said, we also need to be able to separate the good and bad people do. Reiser in his capacity as a programmer is really a different entity to him in his capacity as a murderer. Celebrating one is not celebrating the other.
It is the stereotypical example, but it is such a good one that it deserves to become a cliche. Fritz Haber was sufficiently evil that apparently his wife committed suicide because of him. The Haber–Bosch process is probably one of the greatest technical contributions to humanity, ever, full stop. Maybe the steam engine can compete or a handful of other things. We can't allow technical excellence to be shunned because the technician had horrible problems - the shunning lessens us all. We can afford for bad people to be remembered if the price they pay is to create something of great use.
Reiser's accomplishments are already out there, there's no need to continue with his name on something he hasn't contributed a patch to in a decade, and there's literally no harm in changing the name of a program like this; it's not a binary decision between "remember the person" and "use their work," plenty of people have their work utilized despite no one knowing their name. I think that's fine.
"FooCorp will gimp the 700LC with a 30-watt power limit, ensuring it won't compete with the full-performance 700.".
He's been in prison. I doubt kernel driver writing is encouraged there.
> 'there's literally no harm in changing the name of a program like this; it's not a binary decision between "remember the person" and "use their work,"'
You are literally incorrect in saying there is literally no harm. The harm is this: you're engaging in the "Two Minutes Hate." I don't want to live in the kind of corrupt society that rushes out to scrub a person's name from the history books just because a skeleton is (literally) discovered in their closet.
Removing Hans Reiser's name from his creation would be equally immoral to putting him to death--because among other reasons, you cannot guarantee that he is guilty with 100% certainty.
I think that the name should stand, so that the community can collectively enjoy the constant whining and bitching from all corners every time the subject comes up, rather than anyone having a technical discussion about boring things like the filesystem's merits.
Zinc, selenium, boron depletion are some particularly serious examples. The vast majority of the world's agricultural soil is zinc deficient, including most of US soil.
This of course leads to deficiencies in our food also, with many foods in the USA being seriously depleted of nutrition, and the corresponding health effects in the populace. It's a slow-motion downward spiral, and existential crisis.
I understand other nutrients like iron are deficient in some of the more serious locales. Fortification of foods with various nutrients has been tried as a stopgap measure, but this problem is only going to worsen as time goes on.
The plain fact is, industrial agriculture is a dead end road, as you allude to in your comment about synthetic nitrogen fertilizer.
Phosphorus (rock phosphate) is an equally major problem, in that it's used way too much, tends to wash out of the soil and cause its own pollution, and is in decreasing world supply also.
Likewise, potassium is essential in large amounts for plant fruiting and flowering. There isn't really enough of it out there in good, usable, convenient forms.
The only answer is to use manure fertilizer, i.e. organics, and the end of industrial farming. Not going to happen without a major disaster occurring first, because the masses aren't known for their deep thinking, and the people running the show aren't known for anything other than total selfishness, and steering nations purposely into disaster for the sake of their own power or profit.
I have no idea how your post got here on a post about ReiserFS, unless it's an HN bug or something.
It's thorny that they decided to keep the name, and not everyone is comfortable visualizing the various things that happened with Hans Reiser.
Just rename it already, please!
filesystem formerly called reiser
That’s not marketing. It’s more like anti-marketing.
anecdote : I asked a young developer near me 'Hey, do you know the Hans Reiser story?'. The reply was 'Who?'. The reply came from someone familiar with the filesystem, and that general line of work.
tl;dr : Reiser is just a name. It was recently the name of a 'bad guy', but it hasn't always been, and won't always be.
p.s. I've lost data with ReiserFS too. I'm glad I don't have any need for it any longer.
It's a filesystem created by and named after a murderer. Just because you're fine with it does not mean everybody is.
If you truly inhabit a healthy, antifragile state of mind, you must realize that many of us are not as enlightened and do have emotional reactions to things like this.
It doesn't mean we are inferior, it just means we are human.
On a more on-topic point: my experience has led me to believe that the Reiser fs (Reiser4) has a narrow use case only when quick operations (seek/read/write) on an inordinate amount (hundreds of thousands+) of small (<1MB) files is required -- and that high performance is absolutely necesaary.
I failed to find any other use in my experiences with it.
I was quite surprised when I initially ran out of inodes my first go-around, as well.
[0] Really, no one cares that much. Whenever the name comes up in less formal circles, obligatroy low-effort "dead wife" jokes surface, but what usually follows is discussion on the filesystem's merits. The outrage is from a vocal minority that I would deduce aren't in any position where choice of filesystem is of their concern.
It seems more likely to me that overly emotional people, regardless of their interest in filesystems, got triggered and wrote a bunch of comments even without clicking the link.
https://news.ycombinator.com/newsguidelines.html
Don't hate the player, don't hate the game; Adapt and move on.
This goes for everyone involved.
Or, you know, separate art from the author.
FiberFS
Also, it would be helpful to have a comparison table with ZFS BTRFS and other others from technical implementation and features
Parole Eligible Date: 01/2020
Current Location: Correctional Training Facility
Admission Date: 09/05/2008
https://inmatelocator.cdcr.ca.gov/Details.aspx?ID=G31008