In this interview, Hans discusses his background and how he came to create Namesys and Reiserfs. He looks back at Reiser3, describing the advantages it had over other filesystems when it was released and its current state. He then explores the many improvements in Reiser4, describing the plugin architecture and its exciting potential for future semantic enhancements.
Interesting, made me really wanna try the FS
I’m a big fan of the reiser, though I didn’t know that the performance of databases on reiserfs wasn’t what it could be. I’m wondering if it’s truly not good (compared to others) or just not what he wants ti to be? I toyed for a while with reiser4 till I learned that it wasn’t ready for 2.6 (at the time). I couldn’t figure out why it kept hosing my system.
I just looked over the -RC1 change-log of 2.6.14 and it doesnt look like its in there.
I dont know about the server side, but on the desktop side is kind of a big issue.
We really havent had a big update in file systems in a while, but our file system requirements have been changing every year, the big thing this year is meta data for files.
Its time to start looking at the desktop side of things, Linux isnt just a server OS anymore. They really need to take a look at the Desktop side of issues now days.
What in the HELL did you just say? What are you talking about?
There hasn’t been a big update in Linux filesystems because there is no need. The common Linux filesystems already support metadata for files (extended attributes) for quite some time now, in a standards compliant way (in contrast to reiser4). btw, 2.6.14-rc1 has support for userspace filesystems (fuse), which is an interesting new feature.
The lack of “need” could be simply attributed to the human nature of adapting.
There never was a need of roads, or wheels or fire. Things just got easier once we had them. Having all three eventually led to the combustion-engine powered transportation….
What I’m trying to say is that even if the “need” isn’t there now, and even if pepole have a hard time thinking “outside the box”, I can guarantee you that in 10 years from now you will recoginze the inclusion of reiser4 (possibly in combination to v9fs) as the “needed” step to evolve beyond the currently 40 years old legacy file system.
We really need a edit function.
the preview didnt show
‘s in the output
Not really. Just read below the ‘Your comment’ text area and you will see only and [i] are allowed.
Maybe we do need an edit my b tags and i tags were rendered as bold and italics exactly as they should have been but I was to dim to follow my own advice
there – I’ve added a filter for *cough* those who don’t read the posting guidelines *cough*
Now the br’s just magically go away.
The only HTML/UBB tags allowed are for bold and italics.
Only bold and italics HTML/UBB tags are allowed.
Looks like reiserfs4 is finally going to make it into the kernel. I’m interesting to know how much got trimmed to get it in, kernel dev has been been fairly hostile. They seem fairly hostile to the idea of a filesystem having its own plugin system. I half feel that they’re intimidated, but they’ve definately realized they’d be supporting and bundling something which does not really support the linux kernel directly. On the other hand, I think reiserfs’s semantics + plugin system is absolutely vital for OSS innovation at large.
I’d be interested to see how much got cut that Hans wanted in the kernel. I know psuedofiles got cut, and they were one of my favorite little features. A lot of examples boast them, it basically means “metadata”
/home/myren/docs/…/tag/futurism
-myren
“I am not as qualified or clever as our competition, and we aren’t as well funded, but we are much more persistent and rigorous. That is not what I wanted to believe would be my contribution to the field when I was a boy, but so it is.”
“A willingness to believe that data indicates that one is wrong, and sometimes perhaps that everyone is wrong, is essential to a scientist. Boys think that being brilliant will make them a great scientist. Men know that, in the words of Sir Francis Bacon “men are imperfect mirrors of the creator”. and that rigor, thoroughness, and a belief in data over consensus are what really matters. I am a blind man with a stick, and my contribution to society is that I ignorantly poke where none have poked before because I am more sure that I am such a fool I’d better check it than anyone else in my field. My only true insight into the field is knowing what a fool I am.”
These comments prove to me that this guy is a genius. He knows his limits and rather than try and prove how clever he is, he admits to being a fool that just has to work hard to make his software the best it can be. He has been fighting to get Reiser4 into the kernel for months and he has faced more political bs than anyone I’ve ever seen before. Linux avocates always pride themselves on the free as in freedom aspect of what Linux stands for. Unfortunately, the powers that be in the Linux World (Kernal Maintainers) have repeatedly told him he’s free to take Reiser4 somewhere else. While some of their complaints have merit (Database performance issues), most of it is petty bickering and in my oppinion a little jealousy, because Reiser4 will without a doubt replace ext2, ext3, Reiser3 and especially XFS when and if it ever get’s into the main tree. If the whole point of open source is ultimately to gather the great ideas of contributers around the world, how come 6 or 7 people working on the Kernel get to decide who’s ideas are worth including in the grand vision. Now I’m all for every ship having a captain, but perhaps it’s time for a group thats a little more impartial to start deciding what gets into the Kernel and what waits. At least Andrew Morten has seen the light.
Feel free to copy any Linux kernel tree you want, call it the Guppetto kernel tree, and include Reiser4. You can be the truly enlightened Kernal (sic) Maintainer that saves freedom for all.
These comments prove to me that this guy is a genius. He knows his limits and rather than try and prove how clever he is, he admits to being a fool that just has to work hard to make his software the best it can be. He has been fighting to get Reiser4 into the kernel for months and he has faced more political bs than anyone I’ve ever seen before. Linux avocates always pride themselves on the free as in freedom aspect of what Linux stands for. Unfortunately, the powers that be in the Linux World (Kernal Maintainers) have repeatedly told him he’s free to take Reiser4 somewhere else. While some of their complaints have merit (Database performance issues), most of it is petty bickering and in my oppinion a little jealousy, because Reiser4 will without a doubt replace ext2, ext3, Reiser3 and especially XFS when and if it ever get’s into the main tree. If the whole point of open source is ultimately to gather the great ideas of contributers around the world, how come 6 or 7 people working on the Kernel get to decide who’s ideas are worth including in the grand vision. Now I’m all for every ship having a captain, but perhaps it’s time for a group thats a little more impartial to start deciding what gets into the Kernel and what waits. At least Andrew Morten has seen the light.
Perhaps you are missing the point. Your title makes it obvious you don’t understand what free software means. So let me put it this way. You are free to fork the kernel and include Reiser4 in your branch.
If Free meant that you had to accept all changes to your branch of a piece of free software that someone suggests, even if you don’t agree with them, then I’d really feel sorry for companies like Red Hat and Novell trying to stabalize their own products.
Anyways, this is besides the point. I just hate that something as simple a concept as Free Software seems to be beyond the grasp of most people’s intelligence on this site.
I’ve bashed reiser4 and Hans previously so it’s only fair that I give him a glowing review here. The previous poster is right. The guy is brilliant. His project is so grand and ambitious, and it has taken so long to get where it is today, yet he, leading a small team of coders, has paved the way to demolishing any and all competition in the filesystem space. He says is best when he said that (to paraphrase) there is no market for the second best filesystem.
The flamefests of the LKML have been… well, they haven’t painted either side in a very good light. The last time I read it Hans was asking for a discrete list of demands from the Linux kernel maintainers and the VFS team, and it seems that work has been done to this end (mentions this in the interview). His point about VFS, standards, and innovation really made a lightbulb go off in my head. Why should VFS remain stagnant when filesystems are evolving? What makes the VFS semantics morally superior to those of Reiser4, such that we can ignore any technical superiority of those of the latter?
I have been bitten bad by Reiser4 several times, and I find it hard to believe, personally, that Hans, ever the scientist, is happy with the results. The two most important machines in my life–my everyday laptop and my mother’s computer that I administer remotely–have both recently suffered corrupted root filesystems with reiser4, and it feels like ext3 with dir_index is faster for desktop and workstation usage.
But the scientist in me realizes that I cannot judge a filesystem based on my limited experiences, and the engineer in me acknowedges the architecture puts the Linux community in a great position to grow into the desktop space in particular. And if I follow Hans’ advice and think of myself as a fool, then it was obviously something I did wrong that led to my bad experiences with reiser4.
As a side note, I *really* thought that this posting would generate more stimulating discussion than it actually did (so far). Did anyone else actually make it through the interview? Or were they looking at the crappy new Windows screenies instead?
What makes the VFS semantics morally superior to those of Reiser4, such that we can ignore any technical superiority of those of the latter?
The most basic problem is compatibility. Even if you change the VFS layer wholesale, if your program uses the new semantics, it suddenly becomes incompatible with every previous version of Linux, along with every other UNIX and every other OS. Innovation and hard work won’t address the compatibility issue — only painful and downright ugly compromise can.
Allow me an analogy. Designing a piece of mainstream software without dealing head-on with compatibility is like designing an airplane without dealing with gravity. No matter how great your design is, no matter how much potential it has to change the world, no matter how beautiful and elegant it is, if it doesn’t deal with good ‘ole Mg, it’s a no-go. Compatibility, like gravity, is a fact of life. Everybody agrees its a problem, but there is no doing anything about it.
The law of compatibility has shaped the computing world as we know it. Software today is not the product of a selection process that retained the best and weeded out everything sub-par. Rather, it is the result of a process that judged every innovation in terms of compatibility. Innovation is a relatively unimportant critereon in comparison.
The most basic problem is compatibility. Even if you change the VFS layer wholesale, if your program uses the new semantics, it suddenly becomes incompatible with every previous version of Linux, along with every other UNIX and every other OS.
What you’re asking is that whatever is available now is forwards compatible with Reiser 4, and you’re simply not going to get that kind of compatibility in any kind of software. Backwards compatibility shouldn’t be, and isn’t, a problem.
If they can make the features and namespacing semantics available to other filesystems if they want it then Linux has the potential to push Unix-like filesystems on to another level and create new standards from the kernel up. That forwards compatibility issue can be negated somewhat in that way. Reiser 4 may be on its own initially, but that will certainly change as other filesystem developers look at what they can start doing. The only problem with compatibility I’ve seen is that a lot of what’s in there is Reiser 4 specific at the moment, but it cannot have anything to do with compatibility with existing filesystems.
Rather, it is the result of a process that judged every innovation in terms of compatibility. Innovation is a relatively unimportant critereon in comparison.
You’re right in a manner of speaking, but totally wrong in the final analysis. Innovation and change is always relatively unimportant until someone else starts doing what you wouldn’t, which then forces the issue. ACLs have been one such area in the filesystem world. Reiser 4 is an opportunity to push things along, and be able to use the right way of doing things to provide absolutely unrivalled performance which no one else can match unless they rip out all of their code and start again. It is a big opportunity, because all of that adds up.
What you’re asking is that whatever is available now is forwards compatible with Reiser 4, and you’re simply not going to get that kind of compatibility in any kind of software. Backwards compatibility shouldn’t be, and isn’t, a problem.
Uhm, I think rayiner had it right because if you write a tool that relies on new features in the VFS, this tool won’t be able to run on older Linux installs, and won’t be portable to other systems such as the BSDs.
What you’re asking is that whatever is available now is forwards compatible with Reiser 4, and you’re simply not going to get that kind of compatibility in any kind of software.
Then Reiser4 (and filesystems like it), will have a very long, uphill battle towards acceptance. Microsoft has it easier, they can force their developers do go in a particular direction. The *NIX folks can’t do that. This is especially true for the desktop guys, who won’t use any tech that can’t be easily ported to MacOS/Windows.
If they can make the features and namespacing semantics available to other filesystems if they want it then Linux has the potential to push Unix-like filesystems on to another level and create new standards from the kernel up.
Linux is in no position to create new standards.
That forwards compatibility issue can be negated somewhat in that way. Reiser 4 may be on its own initially, but that will certainly change as other filesystem developers look at what they can start doing.
That’s assuming that other filesystem developers move in the same direction as ReiserFS. It’s not going to happen. I don’t think you quite understand the magnitude of the changes possible with Reiser4. It’s very blue-sky stuff. Eg: instead of storing an MP3 as a string of bytes, you store it as a directory of files, with one file for the compressed data, one file for each of the ID3 tags, etc. Or, consider Reiser’s example of storing XML files on a per-element basis in the filesystem.
This sort of thing is very powerful, yet completely useless until people write apps to take advantage of this functionality. And people won’t write such apps, because the results won’t be portable to any other operating system. Indeed, the KDE and GNOME folks are moving in a completely opposite direction — moving a lot of filesystem semantics into userspace (via KIO, GNOME-VFS, and D-VFS). If and when KDE and GNOME see search functionality integrated into the desktop, you can bet it will be something in KIO or GNOME-VFS that works on any old filesystem. Sure, it’s a crappy solution that will greatly limit the speed and capability of the system in comparison to something similar implemented on top of Reiser4, but nobody will care. Because compatibility is King.
Innovation and change is always relatively unimportant until someone else starts doing what you wouldn’t, which then forces the issue.
For example?
ACLs have been one such area in the filesystem world.
ACLs are a bad example, because you can get most of the benefits of ACLs without rewriting your software. Very little software utilizes features of ACLs that they don’t get automatically from the OS. A better example is NT’s security model. It’s very flexible and very powerful, and almost nobody uses its full capabilities. Why? Because it’s different! Because if your planning a Mac port for your software, you can’t use it! Microsoft has a little bit more maneuvering room, because many Microsoft developers don’t care about portability, but even they are hitting their limits. The developer response to new tech like Avalon and Indigo has been, well, lukewarm. Have you met anybody that is excited about Indigo? Wonder why? It’s not compatible!
Interestingly – and slightly off topic – Han’s comments on doing a BSD version of his ReiserFS:
“As for BSD licenses, I am not that generous. If other people want to charge for my code, they should give me some too. I do offer licenses in addition to the GPL for a fee.”
Obviously Hans groks the real difference between the GPL and the BSD – ie. what I’ve been saying all along – the BSD license are NOT true open source. They allow those to take it and close it. Once it’s (the src code) is closed, then it’s no longer open. The GPL forces the code to remain open, and that’s why it’s truly “open source” in my eyes.
Dave
Obviously Hans groks the real difference between the GPL and the BSD – ie. what I’ve been saying all along – the BSD license are NOT true open source. They allow those to take it and close it. Once it’s (the src code) is closed, then it’s no longer open. The GPL forces the code to remain open, and that’s why it’s truly “open source” in my eyes.
Oh great. Time for the GPL troll. Hans is not saying anything of the sort. What he is saying is that he doesn’t want anyone to profit from his code while not giving anything back. It has nothing to do with closed or open code.
Gezas, get a life!
Oh great. Time for the GPL troll. Hans is not saying anything of the sort. What he is saying is that he doesn’t want anyone to profit from his code while not giving anything back.
Yer, and that’s what the GPL provides for Hans. Hans is saying that if people and companies use the software then he wants some code and payback. It’s one of the cornerstones of Linux’s success.
It has nothing to do with closed or open code.
Since what he’s talking about is all about code, then yes it is. I suggest you go back and re-read what he’s said, because there is no other interpretation.
Once again the GPL moron shows he’s a complete idiot. Hans Reiser can and has said that he offers his FS code with a different license than GPL.
You GPL drones have the iq of an insect.
Quote: “Hans Reiser can and has said that he offers his FS code with a different license than GPL.”
That’s right – at a *cost* to compensate for the lack of improvements going back into the code. Read and comprehend.
Quote: “Once again the GPL moron shows he’s a complete idiot”
Funny, I’d have thought you were, considering you couldn’t comprehend the basics of Han’s statement. I’d vote your comment down, but I’d rather not waste my time with moronic idiots like yourself.
Dave
Obviously Hans groks the real difference between the GPL and the BSD – ie. what I’ve been saying all along – the BSD license are NOT true open source. They allow those to take it and close it. Once it’s (the src code) is closed, then it’s no longer open. The GPL forces the code to remain open, and that’s why it’s truly “open source” in my eyes.
Did you even read the interview moron? Hans doesn’t have a problem with proprietary software and even talks about balance, as well as driving and old car and being $200k in debt. So he doesn’t agree with your idiiotic philosophy.
As for BSD licenses, I am not that generous. If other people want to charge for my code, they should give me some too. I do offer licenses in addition to the GPL for a fee.
I guess Hans doesn’t understand that people already charge for code he wrote without giving him a dime. I’m sure Novell and RedHat thanks him though.
I guess Hans doesn’t understand that people already charge for code he wrote without giving him a dime. I’m sure Novell and RedHat thanks him though.
SUSE did fund reisferfs and Red Hat doesnt support anything but ext3. so…
Quote: “Did you even read the interview moron?”
Ahh. Such eloquence. Yes, I did. So, because my comments and views (and direct quotations from the aforementioned interview) do not agree with yours, I’m now a moron. How lovely.
Quote: “Hans doesn’t have a problem with proprietary software and even talks about balance, as well as driving and old car and being $200k in debt. So he doesn’t agree with your idiiotic philosophy.”
Did I say Hans had a problem with proprietary software? No. So, please don’t put words into my mouth that I didn’t say. As to being 200k in debt, Hans had several choices. Release under the GPL and ensure that he gets improvements to the src code back, or release under the BSD and let anyone use it and take his hard work, without giving him any credit other than a copyright header notice in the code. Or, he could have even made his own license up, that combines both GPL and BSD elements, and gives him control etc. Nothing is stopping Hans from selling his GPL’d code. If you want to argue with that, I’m sure Redhat would disagree with you. Of course, if all you want to do is exchang insults, and twist things to suit your own precious viewpoint, then go for it.
Quote: “I guess Hans doesn’t understand that people already charge for code he wrote without giving him a dime. I’m sure Novell and RedHat thanks him though.”
The point is semi valid. And only semi. The GPL allows others to sell his code yes, as part of a package etc, ie a Linux distribution. His viewpoint (and the main difference between the GPL and BSD licenses), is the ability to take his code, fork it, make money from the forked changes, and not give those improvements back to him. That’s his real beef in the article (or at least how I perceive his comments). That’s what he’s really against. Not selling reiserfs code as part of a package and others paying for it (and himself not getting a “cut” of the distribution fee).
And, as another person has pointed out, Suse has provided funding to him. I believe Suse has also funded the k3b developer(s). So, even if they’re GPL licensed code, the added benefit of having these hardworking people develop their code/projects isn’t lost on some companies. A BSD style license would simply encourage taking and no giving. I mean, human nature is greedy, why give if you can take and get away with it? Take a look at what SCO has been trying to do, and it’s free lunch spiel. Imagine the fun SCO would have had with the BSD license!
Dave
Ahh! Tell you what, let’s rip out all of the GPL stuff out of the BSDs, see how much you have left.
Hey dickless, the only thing the BSDs rely from the nazis over at GNU is GCC. And those assholes aren’t the only ones that know how to write a C compiler.
And, as another person has pointed out, Suse has provided funding to him.
Yeah, here’s some spare change from my couch Hans. Thanks for the 10 years of work idiot.
Interestingly – and slightly off topic – Han’s comments on doing a BSD version of his ReiserFS:
“As for BSD licenses, I am not that generous. If other people want to charge for my code, they should give me some too. I do offer licenses in addition to the GPL for a fee.”
Obviously Hans groks the real difference between the GPL and the BSD – ie. what I’ve been saying all along – the BSD license are NOT true open source. They allow those to take it and close it. Once it’s (the src code) is closed, then it’s no longer open. The GPL forces the code to remain open, and that’s why it’s truly “open source” in my eyes.
Dave
Why does everything on this site erupt into a license flamewar because a couple pathetic morons that post here can’t drum up the brains to talk about anything else? If you don’t like the BSDL, then fuck off and don’t use it. Quit wasting our time with your stupid bullshit comments that have absolutely no relevance to the topic on hand.
How the hell hasn’t this guy been modded out of oblivion yet? Oh, I forgot, all the mod points are being saved for disagreements, not trolls.
Oh, and to feed the troll just a bit, once the code has been closed by a company it does in-fact stay open. Or are you trying to tell me FreeBSD no longer exists because Apple closed a bunch of the code?
Quote: “If you don’t like the BSDL, then fuck off and don’t use it.”
Ahh! Tell you what, let’s rip out all of the GPL stuff out of the BSDs, see how much you have left.
Quote: “Quit wasting our time with your stupid bullshit comments that have absolutely no relevance to the topic on hand.”
You really don’t have any brains there do you? Firstly – the article is about an interview with Hans Reiser. Read the fucking article heading dickhead. It links to the same kerneltrap.org article that I have read. Since, this osnews.com article links to the kerneltrap.org review, then either the entirety of the review, or parts thereof are fair game to comment on. And since, Hans has made a comment on why his file system is released under the GPL, it was also fair game to quote and post here on osnews.com. So, for all your bullshit about being off topic, you are quite wrong. I’m completely on topic.
Now, if the osnews.com staff would get off their lazy asses and actually mod the stories and comments correctly, we would have a problem. But, since we have a large collection of BSD luddites that post on osnews.com, that are pro active in modding down anything that doesn’t conform to their very limited view of “free software”, we have issues with the current osnews.com modding system I believe. Sure, I could get all the pro GPL goes to mod me up, but what does that prove? It shows that the current system is totally inadequate. There is no reason why any of my previous posts should have been modded down, since, if you look at the osnews.com guidelines I haven’t done anything to warrant being modded down. So, tell me, who is going to actually moderate the abuse of the modding system that is currently in progress? mmm?
Quote: “Oh, and to feed the troll just a bit, once the code has been closed by a company it does in-fact stay open. Or are you trying to tell me FreeBSD no longer exists because Apple closed a bunch of the code?”
Ah. I just knew you’d bring this argument on. Firstly – Apple is under NO agreement to re-introduce improvements etc, back to the FreeBSD crew. None. They do so out of good will, nothing more and nothing less. But, when it’s convenient for Apple, they’ll stop providing improvements back into the community. It can, and it does happen. Of course, like other BSD luddites, you’ll quite happily say “well, it’s up to Apple, it’s their code, and it’s free, so they can do what they want”. And this is where I fundamentally disagree with you and others that support the BSD style license. Tell me, if I took JK Rowling’s Harry Potter characters/themes, and wrote a new series, based on those characters, would that be allowed? No. Even if you take the copyright issues out of the above suggestion, and made that part irrelevant, I’d be very positive that Rowling would like a very nice chunk out of any financial gains made from it, as well as part ownership, since she wrote the original characters/stories. Get my drift now? The BSD analogy continued, you’d have another author just take the Harry Potter characters/themes, write some new stuff, publish it, and make money, and JK Rowling not get a brass razoo. Get it now? Of course, you seem to be incredibly dense (yes, I too can be childish and throw insults around as well), so you most probably won’t get it.
Will the GPL survive? I doubt it, business wants it dead. They’ll choose BSD licensed code any day, because it allows them to bastardise the code to suit their needs, without any necessity to return to the community. Business loves that. Take, and not give back. RMS has the right idea on sharing, and if more humans actually did so, the world would be a lot better place. As it stands, licenses like the BSD license encourage taking others hard work and making some changes, and then selling at a nice profit, to benefit people financially, no matter what the social or moral costs of such irresponsible actions are.
Dave
For those that love the BSD license, and consider the GPL as not being truly an “open” license, then maybe you should consider not using ReiserFS, since it’s designed with the principle of GPL and true openness, and obviously it must conflict with your “ideals”. I mean, if you’re a person of principle, of true principle, you’d not use any GPL’d stuff since you don’t like the license. Of course, you all do, since you like to sponge off GPL stuff.
Dave
Quote: “What he is saying is that he doesn’t want anyone to profit from his code while not giving anything back. It has nothing to do with closed or open code.”
Have IQs dropped on this site? It seems that they have. Or basic comprehension skills. Or both. Yes, definitely both.
The whole idea of the GPL is to make sure that improvements that are made go back to the community. For the community. The BSD license does not ensure this. In fact, it fights vehemently to make sure that this doesn’t happen. And yes, it does have absolutely everything to do with open and closed code. Open code remains open, in it’s entirety. Closed doesn’t. If the ReiserFS had been released under a BSD license, you’d see numerous ripoffs from Microsoft and other corporate leechers, with no improvements going back to the core code that they’ve leeched. And, those same improvements would remain private and closed. That is NOT open source. Far, far, far from it.
Big business loves the BSD license because it can take, and never have to give back. That is greedy, and immoral. Of course, modern business doesn’t really give a fuck about being “moral”, does it? Money is the big god, money and control.
You know, all of you BSD trolls can mod me down till the sky turns pink, it doesn’t change the basic facts and truths of my (and others) arguments. And, I must be striking pretty close to the truth, since it seems to piss off so many BSD Luddites. Oh, and the modding down doesn’t really worry me, it’s just a form of social control to silence people, nothing more and nothing less. So much for true freedom of speech.
Dave
PS Oh, and get the balls to get an account, instead of being an AC.
Quote: “What he is saying is that he doesn’t want anyone to profit from his code while not giving anything back. It has nothing to do with closed or open code.”
Have IQs dropped on this site? It seems that they have. Or basic comprehension skills. Or both. Yes, definitely both.
The whole idea of the GPL is to make sure that improvements that are made go back to the community. For the community. The BSD license does not ensure this. In fact, it fights vehemently to make sure that this doesn’t happen. And yes, it does have absolutely everything to do with open and closed code. Open code remains open, in it’s entirety. Closed doesn’t. If the ReiserFS had been released under a BSD license, you’d see numerous ripoffs from Microsoft and other corporate leechers, with no improvements going back to the core code that they’ve leeched. And, those same improvements would remain private and closed. That is NOT open source. Far, far, far from it.
Big business loves the BSD license because it can take, and never have to give back. That is greedy, and immoral. Of course, modern business doesn’t really give a fuck about being “moral”, does it? Money is the big god, money and control.
You know, all of you BSD trolls can mod me down till the sky turns pink, it doesn’t change the basic facts and truths of my (and others) arguments. And, I must be striking pretty close to the truth, since it seems to piss off so many BSD Luddites. Oh, and the modding down doesn’t really worry me, it’s just a form of social control to silence people, nothing more and nothing less. So much for true freedom of speech.
Dave
PS Oh, and get the balls to get an account, instead of being an AC.
No, you seem to not comprehend why you are getting modded down. This is a story and thread about the ReiserFS Filesystem. You are talking about GPL vs. BSD. Now, when you click to mod someone down, there a set of criteria you should follow before modding someone down. I will post what it says here.
NOTE: This feature is to report comments that are offensive, inflammatory, off topic, or otherwise in violation of the OSNews forum rules detailed on this page.
Like I said in my last post, your crap about GPL vs BSD has absolutely no relevance to the topic on hand. You are simply talking about it to start a flame-war and distract everyone in this thread from the topic on hand, which, in-case you have forgotten again, it is the Reiser Filesystem. Nothing more and nothing less.
Understand now?
No, you seem to not comprehend why you are getting modded down. This is a story and thread about the ReiserFS Filesystem.
You might want to re-read the title of the article: Hans Reiser: Interview. What that means is that it is not just about filesystems, but also the content of that interview as well. You have actually read it I presume? It’s not just about the summary. In that interview Hans Reiser is asked for his opinion on licenses and developing filesystems for closed systems, thus:
“Jeremy Andrews: How important is the GPL to you? Do you have any interest or intention of working on a filesystem for any of the BSD operating systems, or for any closed source operating systems?”
In reply to that he mentioned that in payment, he would want any company or person to pay him something back in the form of new or modified code to make Reiser 4/FS better. That’s why isn’t he isn’t as keen on BSD licensing. Not that he hates it, just that he isn’t keen and he personally doesn’t find it sensible.
Unfortunately, all this means any issues brought up therein that might be uncomfortable for certain people. However, that doesn’t mean we can’t still mod down stupid comments.
Quote: “Understand now?”
And I disagree with your views. So, I guess that makes me right, and you wrong. Of course, since osnews.com has been blatantly anti RMS/FSF/GNU/GPL for quite some time now, and pro BSD, it’s not unexpected. I’ve seen quite a lot of pro GPL guys say this, we’re not all wrong you know.
The osnews.com article was about an interview with Hans Reiser, it links to said interview. Anything in the context of that interview is fair game for comment. If you don’t like that, and like my comments, tough fucking shit. I don’t really care. Anti free speech nazis like yourself can go take a long walk off a short pier for all I really care. I’m going to say my 2 cents worth, and there’s absolutely nothing you can fucking do about it? Comprehende? Or are you too fucking stupid to understand that?
Dave
…and some of us thought you had a point – until you started with the language and the personal insults.
Yeah well, considering I was insulted several times, by several other posters, because I dared air an opinion that was different to theirs, deliberately gang-modded down by a group of pro BSD guys, I think I have a bit of a right to get a bit pissed off. I suggest that you go and complain to osnews.com about innapropriate modding down taking place if you really want to do something useful.
Hey osnews.com crew, why don’t you start tracking modding down requests via IP, and start seeing if it’s the same people modding [myself] down all the time – might prove that these very people are abusing the modding system and modding myself down, not because of a proper reason to mod myself down, but in order to quell my freedom of speech and thoughts. Since the flat view is set to -1, anything modded down past -1 won’t get seen, unless you manually change the flat view of course. This leaves the current modding system open to abuse. Of course, if the osnews.com staff don’t want to actually do things the right way (highly likely that they’re anti GPL/pro BSD), then of course, I’m barking up the wrong tree.
Here’s news for you guys – I’m going to keep posting the same things, have fun modding me down. If you close my account, I’ll open up another. And another. You’re not going to intimidate myself into staying quiet.
Dave
‘Nuff said about the damn licenses already. Get a life people.
Oh yes, because discussing filesystems on a geek message board is so much more “having a life” than discussing licenses? Yesh.
Licenses, legalities, etc, are a fact of life. Discussing them is hardly off-topic with regards to ReiserFS. Yes, it’s all political, but its stupid to be like your average engineer and bury your head in the sand when it comes to politics. In the Real World (TM), ReiserFS’s license is easily just as important as its journaling code…
And I thought I would read about Hans Reiser or ReiserFS in the comments … *sigh*
Can we have more comments about ReiserFS vs and filesystems please,
“Can we have more comments about ReiserFS vs and filesystems please,”
No, I already tried changing the subject from proper use of HTML tags in posts and/or the need for and edit feature. I got one good response (precisely the type I was looking for, concerning the tradeoff of innovation and compatibility in software development), and then the thread descended into a license war.
Since one more off-topic post isn’t likely to significantly impact the quality of this discussion, I have one question for the OSNews staff: is there anything wrong with modding up/down the children of posts that are modded up/down? We see a lot of trolls getting modded down heavily but the responses (which are no better, really) stay near 0. I don’t want to see responses to terrible posts, and I’ll give the benefit of the doubt (at first) to responses to exceptionally good posts.
I read the interview and found it somewhat informative. Unfortunately Hans has a knack for dashing sympathy -he seems determined to win the negative popularity contest. His claim about being a fool is one of the most hollow examples of false modesty I have ever seen-and I say this as someone who would like to see him and his fs succeed: every time he opens his mouth he fuels those opposed to him and his ideas-even worse his personality detracts from his ideas and code.
Rayiner I disagree with your comment regarding compatibility-you state this as if Hans had not taken compatibility into account in his design of reiser4. It is abundantly obvious from the interview that opposite is true-he designed reiser4 in such a way that , in theory, any pre-existing command line program should work without modification with his new fs. In theory what hans has done is address functionality which previously simply did not exist and consequently all programs should work unchanged because the functioning of the programs is orthogonal to the semantic structure of the vfs. However, in practice, as opposed to theory, some command line programs were impacted negatively by the “silent semantic” changes-but these issues *can* be seen as bugs present in those command line apps-because they obviously make assumptions about the fs which they should not (and should not have to) given a clean and clear vfs implementation layer. As it turns out these *bugs* in these command line apps are pivotal to the functioning of Linux and it is not trivial to change them-even though it would make far more sense to do so rather than having an ambiguous fs/vfs implementation which allows for and neccessitates such *bugs*. No one could have known in advance that such *bugs* would surface-at least not until someone actually stressed the vfs layer like reiser4 does. Hans has been heavily criticzed for this but anyone who attempted to stress the vfs to such and extent would surely have triggered the same respons(stress the vfs in this sense means pushing the ill-defined boundary between fs/vfs-which exposes the latent assumptions about the underlying devices present in said programs).
The resolution of this impasse is:
a) get the reiser4 fs into the mainline kernel without the ‘silent semantic changes’ -mark it as experimental-collect lot’s of bug reports, iron out these bug reports make the new fs stable.
b) then a few distro’s need to take risk of promoting reiser4-more users, more bug reports, more stable.
c) slowly begin to replace and rewrite said applications which contain latent assumptions about the underlying fs/hardware-slow process needs to be done in such a way as to not cripple the immese amount of scripts and system tools dependant upon these *bugs*-this process may take upwards of 2 years and requires all of the interested parties to work together to facilitate such.
d) upon completion of this process we will have a far superior vfs(not the least of which will be due to a better defined boundary between vfs/fs/hardware) and a supremely stable and well tested reiser4 fs.
Linux needs this far more than reiser(Hans) does….let’s hope people can make it happen…
That are a few lines in a complete interview that is about the license, where there is more about Mr. Reiser and ReiserFS. And what is the topic in the comments? Reiser(FS)?