For anyone wanting (read-only, ATM) access to ZFS partitions on Linux today, there is a ZFS FUSE driver available.
Yes, that's the typical Apple solution: you can sort of use it, but if you really want to use it, you have to commit to using OS X. It's not a good proposition.
I am pretty confident that a Linux driver for ZFS will emerge, and in the long run, I wouldn't be at all surprised if it ended up being a very common filesystem on Linux systems.
I seriously doubt there will be an independent implementation of ZFS; that work would probably go into ext5. Even if there were (or if ZFS becomes GPL compatible), I doubt it will get much traction: Linux has had more powerful file systems than ext3 for many years, and people choose not to use them. Impressive feature lists don't make a better file system.
No, they just wanted something that could implement "time machine", their backup-done-right proposal.
There are several open source file systems that support time-machine like functionality. Nor is there anything particularly new about "time machine"--it's a well-known approach that works fairly well in some circumstances and not at all in others.
I expect Solaris to be switchd to GPL some time in the GPLv3 era, at which point there won't be a problem porting ZFS to Linux.
Perhaps, or perhaps not. Either way, it won't make a big difference to Linux. Ext3 is actually a good design; its deliberately limited feature set is what makes it good.
What is the rational behind not supporting ext2/3 natively in OSX? No demand? It would make our lives a lot easier...
Apple's attitude towards many FOSS standards vacillates between indifference and hostility.
For example, they view X11 as a dead end and want to actively convert people to using Cocoa; hence, Apple's X11 support sucks and they have no interest in making it better.
OpenDoc compatibility would be natural for Pages and Keynote, but I suspect apart from being a lot of work, they don't want it: Microsoft Word matters commercially, and they likely view OpenDoc as supporting FOSS competition (since NeoOffice is actually quite sweet on MacOS).
As for ext3, I think there are two reasons. First, using ZFS, Apple has a feature-list advantage over ext3 (never mind that that doesn't translate into any real world advantages). Second, I think Apple really doesn't want to make it easy for people two switch between OS X and Linux; despite the bluster, they must recognize that Linux is a serious alternative to OS X, in particular given the current and upcoming releases of Gnome and KDE.
In a nutshell, the company just isn't committed to FOSS, they are just using FOSS when they see a short-term business advantage.
It's pretty ironic that all the features in Vista Pogue is making fun of as having been "ripped off" by Apple were themselves things Apple "ripped off" from others: desktop search, dashboard, and 3D chess. The same is true of many other features Apple marketing likes to brag about: hardware accelerated windowing, metadata file systems, the Cocoa libraries, translucency, WIMP interfaces, etc.
None of this would really matter much: it's fine for companies to copy from each other. The only reason to even discuss it is because Apple's favorite marketing meme is that everybody is "ripping them off". So, Apple and Microsoft, since you're not willing to acknowledge where the ideas you are using actually have come from, can you both please STFU about "innovation" altogether?
Apple could switch to ext3, a proven, high performance file system. Instead, they pick a file system that is going to be maximally incompatible with Linux. Seems to me, Apple wants to be incompatible with Linux.
If it's not defined by the standard, then web pages shouldn't be using it. It is easy to write fully functional, nice looking, 100% standards compliant pages that do not use any non-standardized behaviors or functionality.
I have little bit experience with gcj on Fedora - and I completely disagree with this statement. On Fedora even supplied Eclipse didn't work.
That's not even a compatibility problem, that's a packaging problem specific to Fedora.
In general, gcj generates executables that are small, efficient, and start up fast, a nice change from Sun Java. It also has many useful libraries that Sun Java lacks.
That logic is fallacious, even if the observable universe is a "simulation", then this simulation runs inside a real universe, and we're at the start again figuring out what the universe is.
It could be an infinite regression of simulations.
Furthermore, software running on a computer can't determine the physical laws underlying the computer it's running on unless it is given access to external sensors. Simulations generally aren't given that access.
Yes, indeed. The issue here is that what masquerades as a "theory" actually isn't one, since the part that would make it falsifiable--a solid statistical foundation--is missing. Despite all the math and simulation, this is merely an idea for a theory, not an actual scientific theory.
My main point is that the GP's statement that his installer still worked was amazing to me.
It's not particularly amazing that that works occasionally. Furthermore, there are some platforms/environments that have that level of binary backwards compatibility far more consistently than Windows.
I almost just expect 20 year old source to compile fine
Good luck compiling and running most 20 year old DOS/Windows stuff. A lot of that code is trying to access hardware directly.
, so I didn't get what it had to do with the GP.
The point is that UNIX got the APIs right from the start, while Windows has had to start over several times. When Windows gives you backwards compatibility, it's because they have to make an extra effort to ship and support old junk that they otherwise would want to ship and support anymore.
That's because stealing panties is a classic sign of a real sex offender getting up the courage to do something more serious.
Says who? You? Heck, why don't we start arresting people for thought crimes, then?
In a nation of laws, people get punished for what they actually do, not for some prediction of what they might or might not do in the future. Apparently, you prefer to live in a totalitarian nation, in which the state can charge anybody with absolutely anything if they just so please.
They are sex offenders and there is a high enough instance of such offenders going on to commit more serious offences to warrent classifying them as high risk.
Says who? Paranoid politicians bent on reelection through spreading fear? Police chiefs who want more power?
If you want to trust your children, the most precious thing you have, to someone then you don't want them to have ANY record.
You also have a responsibility: not to turn your children into paranoid imbeciles before loosing them on society, and it looks to me like you're failing. Besides, under what circumstances do you have to "trust" your children to anyone? Both my parents were working, but growing up, I don't remember ever being left in the care of any strangers in situations where I could have been abused or harmed. Maybe you're simply a bad parent.
And maybe we should throw constitutional rights out the window for the sake of the children and take away all children from their parents at birth; after all, a large percentage of child abuse and molestation happens at the hands of family members. Think of the children! We need to protect them from this danger!
The specs for the language have been available since day one.
The specs for the language and libraries still are not available, except under a restrictive license. Furthermore, the "specs" aren't really specs at all, they are merely documentation; specs need to be far more complete.
There are a couple of very good alternative implementations (free as in beer)
All non-OSS implementations (IBM, Blackdown, Apple, etc.) are derived from Sun's source code.
and some very mediocre OSS attempts.
Yes: they're mediocre because there are no open specs for Java. In fact, there are no specs for Java at all; there is only a bunch of documentation, a bunch of reference manuals, and a proprietary and expensive test suite.
Its been a constant source of amusement to me that for all the 'we have all the best developers etc etc etc' the 'community' has harped on at Sun for years to open source because basically they have been unable to produce a comparable implementation on their own
They've been unable to produce a comparable implementation of Java for the same reason they have been unable to produce a comparable implementation of Windows: there is no spec, there is no open standard, the documentation is proprietary, and the vendor keeps changing the platform around.
Apart from lack of compatibility with Sun Java, the open source implementations are as good or better than Sun Java.
I love statements like this.
Wow, you're the typical Sun fanboy: when Microsoft relies on proprietary specs and keeps changing their APIs around, you cry foul, but when Sun does the same to the FOSS community, you think everybody should thank them for it.
Maybe open sourcing Java will address some of these issues--provided it's going to happen, not exactly certain given Sun's history of broken promises.
That depends on what market we are talking about. Open sourcing Java will make a MASSIVE difference in terms of Java's appeal to the open source development community.
No, it will merely turn the unacceptable into the unpalatable.
d) As Linux distributions integrate and include Java by default, it will increase the appeal of both Java and Linux on the server side.
Yes, that is likely to happen: Linux distributions will ship with more Java server-side stuff. Nevertheless, PHP-based server apps are still going to be predominant because they're easier to develop and lower footprint.
The problem with Java has only in part been the license; mostly, it has been Sun's stifling control over the platform. As a result, Java has numerous technical problems. Of course, if Java had been an open standard for the past 10 years, there'd be dozens of independent implementations right now. They'd be partially incompatible, and that would be a good thing.
Overall, open sourcing Java was necessary for Sun to remain relevant at all; it will stabilize Java for a little longer, but unless Sun is willing to make some radical changes to the Java platform--including a massive cleanup and pruning of the libraries--Java is past its peak.
The question shouldn't be "does it work with Opera", the question should be "is it standards compliant".
The user should simply send a bug report to both the browser and the site developer. Both developers should then determine whether the problem is with standards compliance of the browser or of the site, and whatever is broken should get fixed. "Don't use Opera" and "every site must work with Opera" are both unreasonable principles.
Seriously, you are talking about two seperate things. Double-clicking an Installshield wizard and having it work in Vista when it was COMPILED on a Windows NT 3.51 box is not the same thing as 20 year old source code still working
You must be kidding... you don't seriously think that that works most of the time. These days, a lot of software breaks with a service pack upgrade. If anything, systems like Linux, Solaris, and AIX provide better binary backwards compatibility. In fact, the Linux package managers will not only be able to install 10 year old software, they'll also be able to tell you whether it will work.
See, the nice thing is that with the Linux distribution model, even though they get it, people rarely need to take advantage of binary backwards compatibility, since almost any software they use and install just gets repackaged fresh for the new OS and upgrades without even having to think about it.
Or, to put it differently, when you say "install", you're already admitting failure. For Linux, 10 year old software just work.
Unlike you, I've given plenty of reasons in previous postings.
As it turns out, our interest seems more than warranted [referring to SCO]. There are enough similarities here that "shut up and mind your own business" just isn't going to cut it.
SCO is working itself out. Why? Because it didn't matter what deals SCO cut with anybody, what mattered was that the source code was GPL'ed and that its developers were careful not to infringe. The SCO example undermines your argument that there is anything special to be done prior to some company actually filing a lawsuit.
I can always discuss the deal on Slashdot.
You aren't "discussing" anything, you're just spreading FUD. You have failed to make a single constructive suggestion for what the open source community should do beyond what it is already doing.
Really? To whom, and on what grounds do they choose not to so concern themselves?
It's a legal agreement between Microsoft and Novell and therefore binding only on them, nobody else.
If two thieves make a pact to rob my home and steal my goods, the details of their discussion still matter to me,
That analogy doesn't work here. Besides, the only thing of any consequence you can do is not buy SuSE/Novell products, which I suspect you are already not doing.
So if it's all the same to you, I think I'll be the judge of what matters to me, thank you.
Well, you can get hysterical about whatever you like, but don't expect other people not to call you on your irrational hysteria.
A third of a billion for a promise not sue random windows users, and some PR? Do behave.
Microsoft does a lot of stupid things with their money. I neither know nor care what they were thinking this time because it frankly doesn't matter.
Uh, no, Windows is not good at backwards compatibility: the entire OS has changed radically over the last 15 years. There are some compatibility hacks, but old software does not work well on newer versions of Windows.
In contrast, 20 year old UNIX software compiles, runs, and takes full advantage of modern hardware; the APIs have hardly changed because UNIX got them right in the first place. That includes the window system.
Slashdot Mirror
For anyone wanting (read-only, ATM) access to ZFS partitions on Linux today, there is a ZFS FUSE driver available.
Yes, that's the typical Apple solution: you can sort of use it, but if you really want to use it, you have to commit to using OS X. It's not a good proposition.
I am pretty confident that a Linux driver for ZFS will emerge, and in the long run, I wouldn't be at all surprised if it ended up being a very common filesystem on Linux systems.
I seriously doubt there will be an independent implementation of ZFS; that work would probably go into ext5. Even if there were (or if ZFS becomes GPL compatible), I doubt it will get much traction: Linux has had more powerful file systems than ext3 for many years, and people choose not to use them. Impressive feature lists don't make a better file system.
No, they just wanted something that could implement "time machine", their backup-done-right proposal.
There are several open source file systems that support time-machine like functionality. Nor is there anything particularly new about "time machine"--it's a well-known approach that works fairly well in some circumstances and not at all in others.
I expect Solaris to be switchd to GPL some time in the GPLv3 era, at which point there won't be a problem porting ZFS to Linux.
Perhaps, or perhaps not. Either way, it won't make a big difference to Linux. Ext3 is actually a good design; its deliberately limited feature set is what makes it good.
What is the rational behind not supporting ext2/3 natively in OSX? No demand? It would make our lives a lot easier...
Apple's attitude towards many FOSS standards vacillates between indifference and hostility.
For example, they view X11 as a dead end and want to actively convert people to using Cocoa; hence, Apple's X11 support sucks and they have no interest in making it better.
OpenDoc compatibility would be natural for Pages and Keynote, but I suspect apart from being a lot of work, they don't want it: Microsoft Word matters commercially, and they likely view OpenDoc as supporting FOSS competition (since NeoOffice is actually quite sweet on MacOS).
As for ext3, I think there are two reasons. First, using ZFS, Apple has a feature-list advantage over ext3 (never mind that that doesn't translate into any real world advantages). Second, I think Apple really doesn't want to make it easy for people two switch between OS X and Linux; despite the bluster, they must recognize that Linux is a serious alternative to OS X, in particular given the current and upcoming releases of Gnome and KDE.
In a nutshell, the company just isn't committed to FOSS, they are just using FOSS when they see a short-term business advantage.
It's pretty ironic that all the features in Vista Pogue is making fun of as having been "ripped off" by Apple were themselves things Apple "ripped off" from others: desktop search, dashboard, and 3D chess. The same is true of many other features Apple marketing likes to brag about: hardware accelerated windowing, metadata file systems, the Cocoa libraries, translucency, WIMP interfaces, etc.
None of this would really matter much: it's fine for companies to copy from each other. The only reason to even discuss it is because Apple's favorite marketing meme is that everybody is "ripping them off". So, Apple and Microsoft, since you're not willing to acknowledge where the ideas you are using actually have come from, can you both please STFU about "innovation" altogether?
Apple could switch to ext3, a proven, high performance file system. Instead, they pick a file system that is going to be maximally incompatible with Linux. Seems to me, Apple wants to be incompatible with Linux.
If it's not defined by the standard, then web pages shouldn't be using it. It is easy to write fully functional, nice looking, 100% standards compliant pages that do not use any non-standardized behaviors or functionality.
I have little bit experience with gcj on Fedora - and I completely disagree with this statement. On Fedora even supplied Eclipse didn't work.
That's not even a compatibility problem, that's a packaging problem specific to Fedora.
In general, gcj generates executables that are small, efficient, and start up fast, a nice change from Sun Java. It also has many useful libraries that Sun Java lacks.
That logic is fallacious, even if the observable universe is a "simulation", then this simulation runs inside a real universe, and we're at the start again figuring out what the universe is.
It could be an infinite regression of simulations.
Furthermore, software running on a computer can't determine the physical laws underlying the computer it's running on unless it is given access to external sensors. Simulations generally aren't given that access.
Yes, indeed. The issue here is that what masquerades as a "theory" actually isn't one, since the part that would make it falsifiable--a solid statistical foundation--is missing. Despite all the math and simulation, this is merely an idea for a theory, not an actual scientific theory.
My main point is that the GP's statement that his installer still worked was amazing to me.
It's not particularly amazing that that works occasionally. Furthermore, there are some platforms/environments that have that level of binary backwards compatibility far more consistently than Windows.
I almost just expect 20 year old source to compile fine
Good luck compiling and running most 20 year old DOS/Windows stuff. A lot of that code is trying to access hardware directly.
, so I didn't get what it had to do with the GP.
The point is that UNIX got the APIs right from the start, while Windows has had to start over several times. When Windows gives you backwards compatibility, it's because they have to make an extra effort to ship and support old junk that they otherwise would want to ship and support anymore.
That's because stealing panties is a classic sign of a real sex offender getting up the courage to do something more serious.
Says who? You? Heck, why don't we start arresting people for thought crimes, then?
In a nation of laws, people get punished for what they actually do, not for some prediction of what they might or might not do in the future. Apparently, you prefer to live in a totalitarian nation, in which the state can charge anybody with absolutely anything if they just so please.
They are sex offenders and there is a high enough instance of such offenders going on to commit more serious offences to warrent classifying them as high risk.
Says who? Paranoid politicians bent on reelection through spreading fear? Police chiefs who want more power?
If you want to trust your children, the most precious thing you have, to someone then you don't want them to have ANY record.
You also have a responsibility: not to turn your children into paranoid imbeciles before loosing them on society, and it looks to me like you're failing. Besides, under what circumstances do you have to "trust" your children to anyone? Both my parents were working, but growing up, I don't remember ever being left in the care of any strangers in situations where I could have been abused or harmed. Maybe you're simply a bad parent.
And maybe we should throw constitutional rights out the window for the sake of the children and take away all children from their parents at birth; after all, a large percentage of child abuse and molestation happens at the hands of family members. Think of the children! We need to protect them from this danger!
You mean like preventing MSFT from writing their own incomptable version?
Microsoft has written their own incompatible version.
How long have you been following java for?
I've been a Java developer since 1996.
What exactly has Sun been stifling? Name a couple of things.
GUI toolkits, numerical support, 3D graphics libraries, to name just a few that matter to me.
Wha might be a good thing for you (how so, btw?)
It would be a good thing because it would mean that the market, rather than some engineer at Sun, decides what people need.
would be a nightmare for me and my customers.
And in what way do you think it would be a "nightmare"? You're just fear-mongering.
The specs for the language have been available since day one.
The specs for the language and libraries still are not available, except under a restrictive license. Furthermore, the "specs" aren't really specs at all, they are merely documentation; specs need to be far more complete.
There are a couple of very good alternative implementations (free as in beer)
All non-OSS implementations (IBM, Blackdown, Apple, etc.) are derived from Sun's source code.
and some very mediocre OSS attempts.
Yes: they're mediocre because there are no open specs for Java. In fact, there are no specs for Java at all; there is only a bunch of documentation, a bunch of reference manuals, and a proprietary and expensive test suite.
Its been a constant source of amusement to me that for all the 'we have all the best developers etc etc etc' the 'community' has harped on at Sun for years to open source because basically they have been unable to produce a comparable implementation on their own
They've been unable to produce a comparable implementation of Java for the same reason they have been unable to produce a comparable implementation of Windows: there is no spec, there is no open standard, the documentation is proprietary, and the vendor keeps changing the platform around.
Apart from lack of compatibility with Sun Java, the open source implementations are as good or better than Sun Java.
I love statements like this.
Wow, you're the typical Sun fanboy: when Microsoft relies on proprietary specs and keeps changing their APIs around, you cry foul, but when Sun does the same to the FOSS community, you think everybody should thank them for it.
Maybe open sourcing Java will address some of these issues--provided it's going to happen, not exactly certain given Sun's history of broken promises.
No, it will merely turn the unacceptable into the unpalatable.
d) As Linux distributions integrate and include Java by default, it will increase the appeal of both Java and Linux on the server side.
Yes, that is likely to happen: Linux distributions will ship with more Java server-side stuff. Nevertheless, PHP-based server apps are still going to be predominant because they're easier to develop and lower footprint.
The problem with Java has only in part been the license; mostly, it has been Sun's stifling control over the platform. As a result, Java has numerous technical problems. Of course, if Java had been an open standard for the past 10 years, there'd be dozens of independent implementations right now. They'd be partially incompatible, and that would be a good thing.
Overall, open sourcing Java was necessary for Sun to remain relevant at all; it will stabilize Java for a little longer, but unless Sun is willing to make some radical changes to the Java platform--including a massive cleanup and pruning of the libraries--Java is past its peak.
The question shouldn't be "does it work with Opera", the question should be "is it standards compliant".
The user should simply send a bug report to both the browser and the site developer. Both developers should then determine whether the problem is with standards compliance of the browser or of the site, and whatever is broken should get fixed. "Don't use Opera" and "every site must work with Opera" are both unreasonable principles.
If you have noisy data and you keep analyzing it enough, you'll eventually find some bizarre model that fits it better than a more plausible model.
It's probably best not to have a firm opinion on the shape of the universe until a lot more data is in.
Seriously, you are talking about two seperate things. Double-clicking an Installshield wizard and having it work in Vista when it was COMPILED on a Windows NT 3.51 box is not the same thing as 20 year old source code still working
You must be kidding... you don't seriously think that that works most of the time. These days, a lot of software breaks with a service pack upgrade. If anything, systems like Linux, Solaris, and AIX provide better binary backwards compatibility. In fact, the Linux package managers will not only be able to install 10 year old software, they'll also be able to tell you whether it will work.
See, the nice thing is that with the Linux distribution model, even though they get it, people rarely need to take advantage of binary backwards compatibility, since almost any software they use and install just gets repackaged fresh for the new OS and upgrades without even having to think about it.
Or, to put it differently, when you say "install", you're already admitting failure. For Linux, 10 year old software just work.
And once again the missing word is "because".
Unlike you, I've given plenty of reasons in previous postings.
As it turns out, our interest seems more than warranted [referring to SCO]. There are enough similarities here that "shut up and mind your own business" just isn't going to cut it.
SCO is working itself out. Why? Because it didn't matter what deals SCO cut with anybody, what mattered was that the source code was GPL'ed and that its developers were careful not to infringe. The SCO example undermines your argument that there is anything special to be done prior to some company actually filing a lawsuit.
I can always discuss the deal on Slashdot.
You aren't "discussing" anything, you're just spreading FUD. You have failed to make a single constructive suggestion for what the open source community should do beyond what it is already doing.
It's kind of ironic that China should restore free enterprise and free market competition by providing an alternative to the artificial DVD oligopoly.
Really? To whom, and on what grounds do they choose not to so concern themselves?
It's a legal agreement between Microsoft and Novell and therefore binding only on them, nobody else.
If two thieves make a pact to rob my home and steal my goods, the details of their discussion still matter to me,
That analogy doesn't work here. Besides, the only thing of any consequence you can do is not buy SuSE/Novell products, which I suspect you are already not doing.
So if it's all the same to you, I think I'll be the judge of what matters to me, thank you.
Well, you can get hysterical about whatever you like, but don't expect other people not to call you on your irrational hysteria.
A third of a billion for a promise not sue random windows users, and some PR? Do behave.
Microsoft does a lot of stupid things with their money. I neither know nor care what they were thinking this time because it frankly doesn't matter.
Uh, no, Windows is not good at backwards compatibility: the entire OS has changed radically over the last 15 years. There are some compatibility hacks, but old software does not work well on newer versions of Windows.
In contrast, 20 year old UNIX software compiles, runs, and takes full advantage of modern hardware; the APIs have hardly changed because UNIX got them right in the first place. That includes the window system.
Without committing to anything that they're willing or able to disclose.
If they are not disclosing it, then it doesn't matter.
Microsoft will have had something in return for that money.
Yes: some patent licenses, a PR opportunity, and, most importantly, people like you spreading FUD.
Seems like you're helping them make their investment pay off.