Post a Comment
I can't figure out why anyone would want to buy Namesys. It has no brand recognition outside of its DARPA sponsorship, it doesn't have a lead developer, and the medium-term prospects for Reiser4 uptake continue to be grim.
The only reason to buy an OSS firm is to acquire its human and/or capital assets. Reiser4 has great potential, but you're not buying Reiser4: anyone can take up its development and sell commercial plugins or support services.
If Namesys came with Hans Reiser, and if Hans Reiser wasn't in prison awaiting trial, then it might be a good investment. Without Hans, Namesys has nothing.
I agree with some of your points as Reiser4 could simply be forked and carried on under the GPL. As mentioned in the article though the company also own some proprietry software (such as a file compressor) that could not be forked.
Hans also isn't the lead programmer (at least according to the namesys website) so development on the FS could continue.
I would say though, that ReiserFS does indeed have some brand recognition, particularly in techie circles, and if the benchmarks for Reiser4 are to be believed then it would seem a good chance to own a good file system at a good price...
Edited 2006-12-22 12:27
Not much of an interview. The sum total of what he actually says is:
"I just keep thinking that I'm stuck in George Orwell's 1984," he said while shackled, dressed in a red, prison-issue jumpsuit, and locked in a cement room. "The government has taken away my kids, invaded my house, held me in solitary confinement and caused me all sorts of financial problems."
Well, Hans, you are the prime suspect in a murder investigation, and rightfully so. You've been charged, and rightfully so. Somehow, I just don't see the injustice. Then again, I suspect you're guilty as sin.
That said, this is probably the best possible news for Reiser4. The new owner would likely be much easier to work with, not go out of their way to antagonize the LK developers, *FIX* Reiser4 where it is needed, *rename* it, get it included into the mainline kernel, and make an honest living supporting it and adding custom features.
Congratulations to all of you who care about that filesystem... except for Hans, of course.
Well, Hans, you are the prime suspect in a murder investigation, and rightfully so. You've been charged, and rightfully so. Somehow, I just don't see the injustice. Then again, I suspect you're guilty as sin.
..guilty until proven innocent? Any proof of your assertions?
--Robin
"Innocent until proven guilty" is an excellent policy for the law and for the courts. Holding individuals' opinions to it is just silly. Circumstantial evidence is perfectly good evidence, especially when there is a *lot* of it. However, in a legal conviction, there really should be no doubt.
My personal opinion will not convict Mr. Reiser, and I am perfectly free to come to a reasonable conclusion based upon the evidence available. And it looks to me like he is guilty as sin.
yep.
I had it in a machine with 3 hard drives.
drive one
/dev/hda1 /boot 100mb ext2
/dev/hda5 / 20gb reiserfs
/dev/hda4 swap 1gb swap
/dev/hda3 /win 96gb vfat
drive two
/dev/hdb1 /home 160gb xfs
drive three
/dev/hdd1 /storage 80gb ext3
/dev/hdd3 /var 20gb reiserfs
there was a power failure, and when I restarted, the machine was throwing up errors on the / and /var filesystems. I used a live-cd to have a look around and could not access the / partition at all. /var was full of weird characters, but all the other filesystems held up and lost no data.
I usually use XFS if I want speed, but at the minute, using Ubuntu, I a, using ext3 on /
with data=ordered as the option.
not the fastest, but very secure.
"""I am using ext3 on / with data=ordered as the option."""
And I thought you were just joking about Reiserfs trashing your machine. I'll bring the rope! ;-)
Agreed that Ext3 is rock solid. I would recommend that you actually benchmark it against other filesystems doing things that you normally do. Ext3 is quite fast. And where it is less so, the difference is not really that significant.
The exception being when there is a nasty interaction with firmware or drivers. I've heard from a couple of people who were experiencing horrid performance with a particular model of 3Ware card and ext3. XFS happened to work quite well in those scenarios.
"Innocent until proven guilty" is an excellent policy for the law and for the courts. Holding individuals' opinions to it is just silly. Circumstantial evidence is perfectly good evidence, especially when there is a *lot* of it. However, in a legal conviction, there really should be no doubt.
First, I'm not judging if Reiser did it or not. Couldn't care less. That's for others to decide.
Your logic sounds great until you're on the other side of the fence. A lot of circumstantial evidence can be gathered on any topic. And in Hanses case? He was living in that house (who doesn't argue at least few times, and especially if divorce is in question?). He was driving that car (hell, I seriously cut my self few times in my car, meaning if I disappear there will be dna evidence in it) He was never what you would call "people person", more like "dissociative arguing eccentric", so he's bound to look strange when his personality is taken to question. Another example, I bought few books about perfect murder too, not because I would be interested how should I do it, but rather because I was more intrigued by puzzle of algorithm which would cover all the facts.
I seem to fit your profile, but am I guilty?
It is like SCO case, except that for difference from SCO case where they say "We know you did it, now tell us how" here "He seems strange and this is why we suspect..." drives the court case.
It is hard to go against circumstantial evidence. Facts you can fight, global descriptions? much harder
Edited 2006-12-22 14:12
Perhaps I was not clear enough. What part of:
"""
"Innocent until proven guilty" is an excellent policy for the law and for the courts
"""
is so hard to understand?
My opinion doesn't mean diddly squat to the court. And I would not want it to.
Finchwizard posted this good link earlier in the thread:
http://tinyurl.com/yhqqge
Hans' strangeness is irrelevant.
Now, could we possibly talk about how this affects Reiser4?
Edited 2006-12-22 14:26
Now, could we possibly talk about how this affects Reiser4?
Agreed:)
Nothing really, usual OSS policy. Either people will pick up where Namesys left or not. Reiser4 was GPL licensed.
What Reiser wants to sell is company resources and customer relations, not filesystem. One buying off his company would also get his customers and that is what Reiser hopes to sell.
Perhaps I was not clear enough. What part of:
"""
"Innocent until proven guilty" is an excellent policy for the law and for the courts
"""
is so hard to understand?
Actually, it's the statement "Somehow, I just don't see the injustice." that doesn't parse well for your argument.
The injustice is easy to see, if he turns out not to have been guilty after all. You've already judged him and sentenced him based on your opinion.
Yeah, he looks guilty to me as well, but until the court actually proves that to be the case, I'm not going to blithely judge that his current incarceration is justice. I know first-hand what it's like to be jailed on false, trumped-up charges, and I wouldn't wish it on anyone.
Unseen injustice is just as unjust as that which you do see.
What's up with modding people down for voicing their own opinion? This post wasn't off topic or inflammatory in any way. There's no excuse for modding it down. I happen to agree with sbergman27 on this one, but even if I didn't I like to think I have a bit more discretion with the mod down button.
Saying that somebody is as guilty as sin without anything upon which to found one's statement is nowhere close to voicing an opinion: that's identical with making an accusation and defaming a man who has not yet been found guilty and condemned. In a private tête-à-tête conversation all this may be an inconsequential incident, but I'm not so sure it is in a public forum.
> Well, Hans, you are the prime suspect in a murder investigation, and
> rightfully so. You've been charged, and rightfully so. Somehow, I just
> don't see the injustice. Then again, I suspect you're guilty as sin.
Well... can anyone here translate the german word "Stammtischniveau"? That's the only thing that comes to my mind when I read such nonsense.
So he is rightfully the prime suspect... because he is the prime suspect?
He is charged rightfully... without proof? Right, who needs proof when we already have a scapegoat.
It doesn't matter if he is guilty or not. Without *proof* that he is guilty, he should not be affected. Or better, you could just admit that he does not live in a consitutional state, and stop talking about rightfulness.
sbergman27, you used to post in this forum at a migh higher level than you do now. I wonder if your account got hijacked...
I guess you don't even live some 100 miles close to Reiser's home. But sure, anyone on the internet, as you are, knows first, who is guilt in a murder trial without even the body found.
Sorry, but even "clearer" cases can have a big twist. For example: http://www.truthinjustice.org/Paul-House.htm
Heiser is in a bad position. Sure he seems to be the killer. But you can't know if it's the truth, especially not just because you didn't like him before. And although this is only your opinion, I have to raise the question how much/less evidence you need to form your opinion?
Edited 2006-12-22 12:26
"""But you can't know if it's the truth..."""
Of course not. And I have not claimed that I do. Others have jumped to the conclusion that I have said so. (Ironic, when you think about it, huh?)
I have said that I "suspect" he is guilty. I have said that "it looks to me like he is guilty". I have *not* said that I know he is guilty, bloody sleeping bags in the back seat aside.
Time will tell, I suppose. Or maybe it won't if he hid the body well enough. ;-)
No one has commented on my assertion that this is a good thing for Reiser4, BTW.
It's never good to see someone go through this, but reading some sites it's not looking good.
http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2006/12/21/BAG29N3...
You always want to hope that it's not the case but time will tell that's for sure.
Good people of OSNews,
Let us refrain from judging this man, it is not our plight, nor our right.
It is all too easy to come out jumping on whatever situation that presents itself just because you have something to say about it. I am a strong and devoted proponent to free speech, yet free speech does in itself not warrant us the right to go about and force this on anyone thinking it is the absolute and unquestionable truth.
Fact is, we are nowhere near Hans nor anyone related to this case.
Everyone would be better off if we'd reserve judgement until it is appropriate to provide input.
I'm not saying it is inappropriate to have an oppinion on this matter, i'm saying that relaying these oppinions has been done in an utterly tactless and tasteless manner so far.
People should be treated with respect even when they don't deserve it in your eyes. No one is forcing anyone to admire Hans or anything, just as no one is really asking for anyone's oppinion.
People don't listen, people never listen, they only seen their own truths.
I sincerely wish this case to be investigated fully and this to be done in an honest UNBIASED manner.
To pass unargumented judgement on anyone at this stage is to mock justice and it's ways.
- Kevin
Edited 2006-12-22 14:40
"""
Yeah, burn him!
"""
I believe they use lethal injection in California. However, it appears that Hans may be in luck. According to Wikipedia:
"In February of 2006, a de facto moratorium on capital punishment was enforced in California as the state was unable to obtain the services of a licensed medical professional to carry out the execution of Michael Morales. The occurrence came as a result of an injunction made by the 9th Circuit Court of Appeals which held that an execution could only be carried out by a medical technician legally authorized to administer IV medications, as the lethal injection procedure if wrongfully performed could lead to suffering for the condemned, potentially constituting cruel and unusual punishment."
I absolutely agree with your statement, 100%.
But what if he is not guilty,
what if his wife got killed by somebody else, she simply flew or whatever..., and let us even assume the court to be fair and the case will be investigated fully and in an honest unbiased manner and he gets acquited.
He lost
* his company,
* and therefore his job,
* his kids,
* his reputation,
* ...
it won't help him too much anyway.
"i'm saying that relaying these oppinions has been done in an utterly tactless and tasteless manner so far."
That's for sure!
This is too similar to the Andrew Pakhomov case in Huntsville Alabama. Pakhomov is a laser physicist at UAH and was also having marital problems.
http://www.whnt.com/Global/story.asp?S=5653749&nav=1VPuZHVb
I'd wager that there are more than a few cases where really smart researchers had severe domestic problems with the spouse which led to separation or divorce, but not murder. The key difference here is that both Pakhomov and Reiser were doing research (perhaps secretly) for the Pentagon. Reiser was with DARPA, and Pakhomov the Air Force. It could be that both men became security risks and the DOD decided to set them up using the domestic dispositions as leverage, a crazy but not altogether untenable hypothesis. Given the extreme similarity of these two cases, I've come to the conclusion that it is rather premature to declare Hans Reiser "guilty as sin".
"""why does it need a new filesystem ? reiserfs itself doesn't cut it really, unless you like botched filesystems now and then...."""
Agreed about the botchedness. Hopefully, a new owner, with less "vision" and less attitude would fix the botched parts and get the (long stalled) Reiser4 show on the road.
Personally, I'm more excited about ext4, which is not suffering from such a severe case of impedance mismatch with the rest of the kernel.
So what are hfs(+)'s pros and cons? I'm curious.
It's the filesystem used by OSX and Mac-formatted iPods, so I guess it's useful for accessing Mac disks and iPods. Apart from that, it's a journaled (journalized?) file system. I don't know enough about it to say anything about its performance, though...
"""Apart from that, it's a journaled (journalized?) file system."""
I think it's either "journalatized" or "journalated", but I'm not sure. ;-)
I glanced over at Wikipedia and it seems that it uses B*-trees for directories, among other things. And can handle 16 exabytes per file in up to 16 exabyte volumes. So I'd say people's iPods are probably safe for a while. ;-)
Odd that this FS is not talked about more in Linux circles.
Well, I've had it crash on me so many times that it isn't funny. The cause was probably some dodgy hardware in my old Powerbook, but I never had nearly as much problems with my Linux file systems on the same computer. One time I had to reformat my ext2 / since e2fsck segfaulted on me, but since I could read the FS fine, I lost no data from it. Reiserfsck fixed all problems on the other file systems the few times it needed to. When I had problems with OS X, which was quite often, HFS+ could be completely gone. No chance of recovering anything. I would boot from the install CD, run the disc utility, and it would bark out some error message about B-trees and stop.
HFS+ is a complete POS, it is fairly slow and its journaling is marginal at best. I have had my HFS+ partition corrupted so many times that I am weary of going more than 1 week w/o backups. The best way of avoiding corruption is to disable journaling altogether, but it sure beats the UFS support in OSX, which is slower than watching paint dry.
OSX is a great system, but its filesystem support is marginal at best, and apple trying to enter the server market is a joke until they get a fast and reliable FS to their xservers (nice piece of HW, but OSX Server blows chunks performance wise). Apple is the only company that can get away with case sensitivity being a selling point for their advanced filesystem. I mean what is this the 70s? There are some kernel modules to deal with ext3 partitions in OSX, however I am partial to XFS and it would be nice to see OSX support for XFS.
RaiserFS is nice, but there are 3 other journaled filesystems already available in linux (ext3, XFS, JFS). So no big deal, which I guess it is one of the strengths of this OS, it has so many legs/choices that if one gets chopped of the system can move forward development-wise w/o any problem.



