Journaling doesn't mean that you no longer need to clean the filesystem after a crash. It just means that the filesystem is consistent in some (rather nebulous) sense.
Whether your filesystem is journalled or not, you need to check it after a crash. For XFS, you can use xfs_check & xfs_repair for that purpose. Of course to check your startup volume, you need to boot from a bootable CD or a different partition.
In the Netherlands, there are no ATM fees. It used to be the case until a couple of years ago that you couldn't use bank A's ATM to withdraw money from an account with bank B, but something made them change it so that you can make one withdrawal at an "alien" bank per day. It seems a much better solution than having Joe and his dog plug ATMs into the bank networks...
At there has been some indication that XFS is related to the slab corruption.
You could equally say that there have been indications that PREEMPT is related to slab corruption. Christ, 2.6 still crashes regularly for many people. 2.6 is just not stable. That XFS, as a part of 2.6, *might* also not be stable is merely to be expected.
Lets face it, XFS has alot of cruft still
No, it doesn't. A lot of work has gone into reducing cruft over the last year. And it has paid off.
Generally, when something is not ready for a stable line, it doesn't have enough people using it.
Right. Which is why Linux includes amateur radio support. </sarcasm>
Even linus said recently that he knows very few people that use XFS.
Judging by the replies on lkml, many people seem to want XFS in 2.4. But maybe more people would use it if it finally becomes a part of 2.4.
XFS gets a lot of flak. Most of it completely unfounded and wholly inconsiderate of the hard work done by the fine people at SGI to port XFS to Linux. Never mind that it's wholly inconsiderate to the large body of people relying on XFS daily.
The kernel maintainers have been clear on their reasons *not* accepting xfs into linus's tree
The kernel maintainers haven't been clear on this at all. It's not even true, since XFS has been a part of Linus' tree since 2.6.early.
Considering that so many of the people who clamor for xfs (imx) are kids who're principally attracted to it's rep for high
XFS has been in service with SGI for over a decade. Arguably it has seen more heavy duty action than most every other filesystem in Linux. It's reputation for being "unreliable" is completely unfounded. At the same time another filesystem for which many, many reports of data loss have come across the lkml, viz. ReiserFS, gets merged into the kernel almost as soon as it compiles. But nobody seems to mind that.
I'm sure if SGI actually cleans up the interface it'll go in but who knows if _that_ will ever happen.
This has *always* been the argument against the inclusion of XFS, even though the SGI people have done virtually *nothing but* clean up the XFS interfaces since XFS release v1.0. The facts are that there is little, if anything, remaining to clean up: people just don't like XFS.
Marcello on XFS:... I dont like the XFS code in general.
That is what it boils down to.
The reality is that Marcello has held off on merging 2.4 for over a year, telling the XFS developers to "come back later" or to "clean up X". Now that pretty much everything has been cleaned up, and the definitive deadline is getting near, the real reason for the holdup has finally emerged: Marcello doesn't like the XFS code.
Well, that's his prerogative. But he should have told them in advance. They could have spared themselves the effort!
I have discovered a truly marvelous sig, unfortunately the sig limit is too small to contain i $ echo 'I have discovered a truly marvelous sig, unfortunately the sig limit is too small to contain i' | wc
1 17 95
Before making claims like these you might want to verify that what you're saying is actually true. I've never had Google redirect me to some other site. I've never seen evidence of sites "buying a ranking" from Google. It's the most unbiased search engine out there bar none.
The question is, who are you to step in and police Google? It's not like they owe you anything. And I'm sure you don't need Google to obtain some Ambien.
Google has always returned slightly different results for seemingly identical queries. There was a whole Slashdot article on the subject a while back. This is simply how their huge system works. There's nothing sinister about it.
I really don't see how there is any analogy between an OS crashing and Google not finding what you need. Did you ever consider that maybe the tool is "failing" simply because you picked the wrong tool?
Seriously though, the kind of cash that would be required to buy an election this way is prohibitive, certainly at the federal and state levels.
Not really. You don't need to bribe everyone. You just need to bribe enough. Ten or twenty thousand votes can make the difference, as can 2 or 3 states.
It's not cynicism, it's reality. Money is indisputably a corrupting influence in our elections, regardless of whether we disclose votes or not.
Yes, but that is not the point of contention. If votes can be bought and sold as you were suggesting, then why vote?
If you can't understand libertarianism, just say so. There's no need to get abusive.
I understand libertarianism is a philosophy for white upper middle class sophomores who have never been in need and don't support anybody. You'll grow out of it.
The startup performance hit isn't in the size of the toolkits, its in the dependencies. Both KDE and Gnome need to start server(s) to provide basic naming/lookup services. These servers need to look in various config files, and everything needs to go through various layers of modularization/internationalization crud. So starting an application causes a storm of process forking and disk access, which slows things down considerably.
Personally I feel the principal reason for many of the problems with a lot of the GUI applications written by volunteers isn't to be found in "hard" technical givens such as library file size or scheduler efficiency, but in the "soft" philosophical commitment that many Linux developers have towards writing programs that are as "lazy" as possible, postponing important decisions as long as possible. This expresses itself in highly modular, clean designs that are very powerful and flexible, but whose flexibility doesn't support the needs of everyday practice. To put it bluntly, volunteers enjoy writing frameworks or systems that approach some platonic ideal; they don't like getting bogged down in messy practicalities.
Your proposal to add another layer of abstraction will do nothing to solve the problem -- it will only compound it.
Oracle is neither free nor libre, so it's not a particularly good example of open source software beating proprietary software. It's just a case of one proprietary package beating another, but that's not very interesting.
The issue isn't UNIX vs. Microsoft, it's open source software vs. Microsoft.
Gongrijp knows what he's talking about. He was one of the founders of Hacktic magazine, a "magazine for techno-anarchists" that was published from 1989 till 1994. Hacktic publications included schematics for pay television descramblers, detailed expositions of operating system vulnerabilities, articles on "social engineering" (I think they might even have coined the phrase), and numerous topics on hacking the phone company ("phreaking") and war dialing.
These guys have also organized some huge hacker conferences such as Hacking at the End of the Universe in 1993 and Hacking In Progress in 1997 (I was there in '97). Later Hacktic professionalized and they became the first ISP in the Netherlands. Still later that turned into XS4ALL, probably the best ISP in the Netherlands.
Through everything, Gongrijp ("Public Enemy #1") was a driving force. If he says the phone is secure, then that's a pretty damn strong endorsement.
AI isn't about emulating humans but about matching humans in mental capacity. How it will accomplish that is up to the researchers.
The problem with that concept is that then we might never know whether we're dealing with intelligence or not. So far, the only example we have is ourselves. In order for us to recognize something as intelligence it will have to remind us of ourselves. By your definition, the refrigerator in your kitchen might be extremely intelligent -- it just has no way of letting us know.
It lacks standardization and most popular modern applications are not available for it.
If you define Linux' success by the number of "popular modern apps" available for it, then I'd rather Linux never succeeds.
One of the more interesting things about Linux is that it gives people a chance to resist the slavish dependance on monster apps and to break the software monoculture.
Otherwise -- and you say it yourself -- why not just run Windows?
The law serves to protect businesses from harm that occurs independently of their business models (theft, corruption, racketerring, etc.).
The question is whether copying a DVD constitutes theft -- i.e. whether digital works need copy protection. The law seems to think so, and I tend to agree.
When you buy/rent a dvd/cd/book etc you DO have a right to view/listen/read it without any additional license.
No, you can't lump all these things together. In order to play a DVD, you need to crack the encryption. Legally you need a license to do so. The DMCA requires this.
No it's more like selling you a ticket and then informing you that you'll have to view the feature while standing on your head and no there are no refunds but we'll happily exchange your ticket for another showing where you'll also have to stand on your head.
It's not like that at all. The analogy revolves around the fact that mere capability does not translate into a right. The right needs to be bought separately -- in the case of a movie, you need to buy a ticket, in the case of a DVD, you need a license to crack CSS.
Just because we are strong and always capable of viewing and ripping their products no matter what they do, that doesn't mean we have a right to do so -- just like somebody who is stronger than you doesn't have a right to forcefully take your money off of you. This is how the DMCA protects publishers.
I do think the DMCA is too draconian, but at the same time I do see the need for protection. I can't quickly think of a way to change it so that it becomes less draconian while still providing the same kind of security.
Slashdot Mirror
Journaling doesn't mean that you no longer need to clean the filesystem after a crash. It just means that the filesystem is consistent in some (rather nebulous) sense.
Whether your filesystem is journalled or not, you need to check it after a crash. For XFS, you can use xfs_check & xfs_repair for that purpose. Of course to check your startup volume, you need to boot from a bootable CD or a different partition.
In the Netherlands, there are no ATM fees. It used to be the case until a couple of years ago that you couldn't use bank A's ATM to withdraw money from an account with bank B, but something made them change it so that you can make one withdrawal at an "alien" bank per day. It seems a much better solution than having Joe and his dog plug ATMs into the bank networks...
At there has been some indication that XFS is related to the slab corruption.
You could equally say that there have been indications that PREEMPT is related to slab corruption. Christ, 2.6 still crashes regularly for many people. 2.6 is just not stable. That XFS, as a part of 2.6, *might* also not be stable is merely to be expected.
Lets face it, XFS has alot of cruft still
No, it doesn't. A lot of work has gone into reducing cruft over the last year. And it has paid off.
Generally, when something is not ready for a stable line, it doesn't have enough people using it.
Right. Which is why Linux includes amateur radio support. </sarcasm>
Even linus said recently that he knows very few people that use XFS.
Judging by the replies on lkml, many people seem to want XFS in 2.4. But maybe more people would use it if it finally becomes a part of 2.4.
XFS gets a lot of flak. Most of it completely unfounded and wholly inconsiderate of the hard work done by the fine people at SGI to port XFS to Linux. Never mind that it's wholly inconsiderate to the large body of people relying on XFS daily.
... I dont like the XFS code in general.
The kernel maintainers have been clear on their reasons *not* accepting xfs into linus's tree
The kernel maintainers haven't been clear on this at all. It's not even true, since XFS has been a part of Linus' tree since 2.6.early.
Considering that so many of the people who clamor for xfs (imx) are kids who're principally attracted to it's rep for high
XFS has been in service with SGI for over a decade. Arguably it has seen more heavy duty action than most every other filesystem in Linux. It's reputation for being "unreliable" is completely unfounded. At the same time another filesystem for which many, many reports of data loss have come across the lkml, viz. ReiserFS, gets merged into the kernel almost as soon as it compiles. But nobody seems to mind that.
I'm sure if SGI actually cleans up the interface it'll go in but who knows if _that_ will ever happen.
This has *always* been the argument against the inclusion of XFS, even though the SGI people have done virtually *nothing but* clean up the XFS interfaces since XFS release v1.0. The facts are that there is little, if anything, remaining to clean up: people just don't like XFS.
Marcello on XFS:
That is what it boils down to.
The reality is that Marcello has held off on merging 2.4 for over a year, telling the XFS developers to "come back later" or to "clean up X". Now that pretty much everything has been cleaned up, and the definitive deadline is getting near, the real reason for the holdup has finally emerged: Marcello doesn't like the XFS code.
Well, that's his prerogative. But he should have told them in advance. They could have spared themselves the effort!
Your sig:
I have discovered a truly marvelous sig, unfortunately the sig limit is too small to contain i
$ echo 'I have discovered a truly marvelous sig, unfortunately the sig limit is too small to contain i' | wc
1 17 95
I'm a nerd. It matters.
Before making claims like these you might want to verify that what you're saying is actually true. I've never had Google redirect me to some other site. I've never seen evidence of sites "buying a ranking" from Google. It's the most unbiased search engine out there bar none.
Who's Google to step in and play the police?
The question is, who are you to step in and police Google? It's not like they owe you anything. And I'm sure you don't need Google to obtain some Ambien.
Tainted? They're just screening their advertisers. Everybody does that.
Google has always returned slightly different results for seemingly identical queries. There was a whole Slashdot article on the subject a while back. This is simply how their huge system works. There's nothing sinister about it.
I really don't see how there is any analogy between an OS crashing and Google not finding what you need. Did you ever consider that maybe the tool is "failing" simply because you picked the wrong tool?
You're assuming that the black hats don't already know of the exploit.
Well, you could try another search engine first. Or thumb through the yellow pages. Or call a friend... Whatever.
Running into dead ends really doesn't reflect badly on Google, it reflects badly on you.
I have several dead-end searches every week.
Nobody is forcing you to use Google, you know.
Seriously though, the kind of cash that would be required to buy an election this way is prohibitive, certainly at the federal and state levels.
Not really. You don't need to bribe everyone. You just need to bribe enough. Ten or twenty thousand votes can make the difference, as can 2 or 3 states.
It's not cynicism, it's reality. Money is indisputably a corrupting influence in our elections, regardless of whether we disclose votes or not.
Yes, but that is not the point of contention. If votes can be bought and sold as you were suggesting, then why vote?
If you can't understand libertarianism, just say so. There's no need to get abusive.
I understand libertarianism is a philosophy for white upper middle class sophomores who have never been in need and don't support anybody. You'll grow out of it.
Why shouldn't you be able to sell your vote?
Because then people who have nothing to offer but money, by definition the worst candidates, will come into power.
It's better than someone else selling your vote for you, isn't it?
Cynicism gets us nowhere.
How you arrive at this decision should be entirely up to you.
And neither does fratboy libertarianism.
The startup performance hit isn't in the size of the toolkits, its in the dependencies. Both KDE and Gnome need to start server(s) to provide basic naming/lookup services. These servers need to look in various config files, and everything needs to go through various layers of modularization/internationalization crud. So starting an application causes a storm of process forking and disk access, which slows things down considerably.
Personally I feel the principal reason for many of the problems with a lot of the GUI applications written by volunteers isn't to be found in "hard" technical givens such as library file size or scheduler efficiency, but in the "soft" philosophical commitment that many Linux developers have towards writing programs that are as "lazy" as possible, postponing important decisions as long as possible. This expresses itself in highly modular, clean designs that are very powerful and flexible, but whose flexibility doesn't support the needs of everyday practice. To put it bluntly, volunteers enjoy writing frameworks or systems that approach some platonic ideal; they don't like getting bogged down in messy practicalities.
Your proposal to add another layer of abstraction will do nothing to solve the problem -- it will only compound it.
Oracle is neither free nor libre, so it's not a particularly good example of open source software beating proprietary software. It's just a case of one proprietary package beating another, but that's not very interesting.
The issue isn't UNIX vs. Microsoft, it's open source software vs. Microsoft.
Gongrijp knows what he's talking about. He was one of the founders of Hacktic magazine, a "magazine for techno-anarchists" that was published from 1989 till 1994. Hacktic publications included schematics for pay television descramblers, detailed expositions of operating system vulnerabilities, articles on "social engineering" (I think they might even have coined the phrase), and numerous topics on hacking the phone company ("phreaking") and war dialing.
These guys have also organized some huge hacker conferences such as Hacking at the End of the Universe in 1993 and Hacking In Progress in 1997 (I was there in '97). Later Hacktic professionalized and they became the first ISP in the Netherlands. Still later that turned into XS4ALL, probably the best ISP in the Netherlands.
Through everything, Gongrijp ("Public Enemy #1") was a driving force. If he says the phone is secure, then that's a pretty damn strong endorsement.
AI isn't about emulating humans but about matching humans in mental capacity. How it will accomplish that is up to the researchers.
The problem with that concept is that then we might never know whether we're dealing with intelligence or not. So far, the only example we have is ourselves. In order for us to recognize something as intelligence it will have to remind us of ourselves. By your definition, the refrigerator in your kitchen might be extremely intelligent -- it just has no way of letting us know.
It lacks standardization and most popular modern applications are not available for it.
If you define Linux' success by the number of "popular modern apps" available for it, then I'd rather Linux never succeeds.
One of the more interesting things about Linux is that it gives people a chance to resist the slavish dependance on monster apps and to break the software monoculture.
Otherwise -- and you say it yourself -- why not just run Windows?
5. Just a side note, Fedora isn't exactly the examplar of Linux's efficiency and performance. Go Gentoo!
Tests don't show any significant benefit for Gentoo versus binary packaging.
mail/file/firewall/web/zope
These are very non-demanding uninteresting tasks, though.
I thought "fair use" covers this.
I don't know that it does.
The law serves to protect businesses from harm that occurs independently of their business models (theft, corruption, racketerring, etc.).
The question is whether copying a DVD constitutes theft -- i.e. whether digital works need copy protection. The law seems to think so, and I tend to agree.
When you buy/rent a dvd/cd/book etc you DO have a right to view/listen/read it without any additional license.
No, you can't lump all these things together. In order to play a DVD, you need to crack the encryption. Legally you need a license to do so. The DMCA requires this.
No it's more like selling you a ticket and then informing you that you'll have to view the feature while standing on your head and no there are no refunds but we'll happily exchange your ticket for another showing where you'll also have to stand on your head.
It's not like that at all. The analogy revolves around the fact that mere capability does not translate into a right. The right needs to be bought separately -- in the case of a movie, you need to buy a ticket, in the case of a DVD, you need a license to crack CSS.
Just because we are strong and always capable of viewing and ripping their products no matter what they do, that doesn't mean we have a right to do so -- just like somebody who is stronger than you doesn't have a right to forcefully take your money off of you. This is how the DMCA protects publishers.
I do think the DMCA is too draconian, but at the same time I do see the need for protection. I can't quickly think of a way to change it so that it becomes less draconian while still providing the same kind of security.