I didn't even knew who Steven Gibson was before this post. Russinovich's site (sysinternals.com) is one of the sites you can't stop visiting if you're doing anything with windows, even NT programmers at Microsoft use it, and Microsoft talks about those programs in several support articles. Just because people has know him after he discovered and analized the sony rootkit doesn't means he has never had a "reputation" as an expert.
Laptops running powerful dual-core CPUs eating less than 30W means nothing to you? Dualcore desktops alone are the biggest change in the CPU world since people started having computers in their homes, IMHO.
They'd have a platform that would whomp all over everything currently in the marketplace
Well, Alpha was a high-end CPU designed for servers. Somehow, I doubt it could whomp the portable market? (which is a important part of the computer world these days)
So - suddenly Intel, whose CPUs are the most power-hungry of the whole CPU industry, has a secret to make CPUs that are 2-3x more efficient (200-300%) than the ones from their competitors from now to 15-20 years (switching an architecture is a long-term strategy, not a short-term one)?
It's just me who thinks this is just bullshit? There are low-power 970GX chips that eat less than 30W (maybe they're not as good as centrino, but they're OK, would you risk the long-term strategy of your company for 5-10W?). Stop telling that "G5 are too hot to be put in a laptop".
It's like the "IBM can't provide a 3 GHz CPU" - AMD has been beating Intel with CPUs at frequencies lower than 3 GHz (right now AMD hasn't CPUs at 3 ghz, just a few 2.8 ghz ones); only intel has needed to go beyond of 3GHz to keep competitive and Jobs asks IBM to provide him a 3ghz CPU when he knows that 2.8 Ghz opterons are beating the 3.8 ghz intel monsters? I guess it's much easier to convince apple zealots that the intel move is the "right thing" if you hide the real reason and invent false excuses, right?
It's not just Intel. Until recently, +95% of the x86 server market did belong to Intel - that means that mobo/chipset/other products makers have their product offerings oriented to Intel. AMD made a great chip, but moving the whole x86 server market from one CPU vendor to another is not something that will happen from one day to another - Intel can't influde that, lots of companies are just not switching to another CPU vendor just because they've had good sales in 1 year. There're no good server-oriented chipsets for AMD chips except the ones from AMD for example (please don't bother mentioning that company that makes graphics cards and started making chipsets yesterday); however you can find chipsets from serverworks or IBM for Intel CPUs. I'm not suprised Intel keeps selling more than AMD, AMD will have to _keep_ the lead a few years if they want to get support from the industry.
I don't know others, but I haven't hit a page that I can't use with firefox in months. I know there're banks (which I don't use) and other stuff, but I think it's fair to say that the number of pages which doesn't work in firefox are 1) unmaintained (and hence, not worth of visiting) 2) a rare exception
What I do care about is that webmasters can't write pages use standars because IE doesn't supports it. Pages not rendering correctly in firefox are sad, but it's also bad that firefox supports lots of things that will never used in websites in a two years timeframe.
Read between the lines Jobs spoke and you find that Apple actually was struggling to come up with convincing evidence that justifies the Intel switch. Nonetheless, Jobs used wording that implied massive (2-4X!) improvements in performance, which, surprise surprise, are clearly massive exaggerations and, in the context of comparing this year's Pentiums to last year's G5s, are actually covers for practical failure on their side....not to mention the "IBM hasn't been able to deliver a 3 GHz chip" crap. Only Intel needed to go beyond 3 GHz to remain competitive, AMD has been the performance/MHz performance/watt leader for more than a year, and they didn't need to make 3 Ghz chip to do that (right now I think that AMD has not even released a 3 Ghz chip, just a 2.8 one) In other words, CPus didn't need to have more than 3 Ghz to remain competitive, only the crappy presscot Intel chips, and Jobs (who has spent years telling people that MHz are not important) knows this very well. He knew that G5 didn't need to go beyond 3 ghz to perform well
Still he promised a 3 Ghz CPU *two*years*before* announcing the intel switch (june 2003). To me, it looks like the Intel switch was decided some years ago, and Jobs said that crap just to have Yet Another excuse to justify the Intel switch to apple zealots. Of course, it has been a succesful move - Apple zealots don't like thinking, and they won't stop thinking twice how stupid the "IBM hasn't been able to deliver a 3 ghz CPU" excuse is.
Speaking of EFI, linux supports it. I've heard lot of noise about windows support, but what about linux support, will linux (and BSDs) be able to run on those things? I'm not buying a machine that can't run open operative systems...
Repeat after me: Mac OS X is not a microkernel. Mac OS X is not a microkernel
Mac OS X is derived from a microkernel (Mach) - but it's far from being a microkernel. A microkernel is not supposed to have the whole TCP/IP stack in kernel space. A microkernel is not supposed to implement drivers in kernel space. A microkernel is not supposed to have the filesystem in kernel space. Microkernels were, in fact, invented to get these things out of kernel space and run as userspace, etc. Being a pure microkernel implied a performance penaly they were not willing to pay. By moving all those things to kernel space, Mac OS X broke the whole "idea" behind microkernels and stopped being a microkernel (which is not a bad thing: all the other OSes do the same).
NT was also derived from Mach BTW. I will never understand why Mac zealots spend countless hours saying how crap the NT kernel is when the fact is that they derive from the same idea. And just because they are derived from a microkernel doesn't mean the microkernel is who implements all the funtionality. Mac OS X and NT copy from mach the "design": processes implementing funtionality runing in userspace, the real kernel being a scheduler scheduling those processes, etc. That's the "framework", the real functionality (TCP/IP, drivers, FS, etc) is implemented on top of that. The difference is that NT implemented that funcionality looking at VMS, and Mac OS X/nextstep implemented it by using BSD code from FreeBSD. There're some exceptions i think (mac os x vfs and journaling layer is implemented by Mach i think) but the idea is that. And this is BTW the reason why Linus Torvalds just hates mac os x and NT design: If you're going to do a microkernel, do it, but if you are going to put all the drivers and tcp/ip stack in the kernel, then don't use a microkernel design base because you're using the wrong tool and you're overengineering, just use a monolithic kernel and modularize it as much as you can (or something like that)
Bullshit. What last.fm does precisely is the same, and I don't see why this is bad. Take a look at my profile if you want to "spy" me. Of course, this is itunes and itunes is "good" and apple is "cool", so some people just feel the need to spread FUD in the hope they "free" people from itunes or "show people the reality behind itunes" or some crap like that - I call that "the superman syndrome"
The WMF format was created for windows 3.1, they may not work at microsoft anymore.
And the WMF "vulnerability" was NOT a vulnerability. It was a feature (you could attach executable code in the WMF for some reason, yes). Of course it was created when there was no internet, but it was a feature and it might have been a useful feature at that time.
It's unusual because Microsoft ALWAYS releases patches on thursdays. People has been asked for years to release them as soon as they're ready, and this time they had to release them sooner, because there were too many risks. The WMF vulnerability has been indeed unusual.
Oh, but this is free software and people packages things in several formats. Windows software is bundled with a _single_ installer and you can't change it.
One of the advantages of windows is that it comes preinstalled and users don't need to partition their hard disks. Linux distros need to force users to create partitions because there's no room for linux by default. Users also need to create a partition of the don't have room for XP. Partitions is not something windows can avoid.
To start with, you've to install tons of apps that the operative systems don't includes itself. And due to that stupid microsoft rule that existed for years ("installer must be executables delivered by 3rd party apps") I've no way to automate the download and installation of those (yes, I know about msi, I also know MSIs can be slipped in the installation CD. I still find no way of installing AND automatic its update like apt-get update & upgrade does. And LOTS of installers are not using MSI still. Shame on you microsoft, for forcing people to create docens of different, incompatible, buggy, installers)
Is it just me or does this sound like a ridiculously ugly approach to data integrity?
It must be just you. By sync'ing your data every 5 seconds they ensure that you can't lose more of 5 seconds of work. Other filesystems try to avoid syncing as hard as possible and you could lose 10, 20, 30 seconds of work...
it appears that ext2/3 wins every single test that matters to me: find dirs, files, untar, tar, copy tarballs(s), the kernel source tree (ext2/3 now outperforms reiser3
It still loses on others. benchmarking it properly can change things...
Distros like Fedora that mandate the initial install ONLY use Ext3 are being stupid
It's amazing that such commentaries are moderated interesting these days. So, uh, fedora developers are stupid and you're smarter than them?. Please take a look at this commentary to understand why such decisions aren't so simple. You can tune your car's engine and it'll be faster, right? But why not everybody tunes their engines?
Let me quote a ext3 paper: "The ext2 and ext3 filesystems on Linux are used by a very large number of users. This is due to its reputation of dependability, robustness, backwards and forwards compatibility, rather than that of being the state of the art in filesystem technology."
I'm *sick* of reading filesystem benchmarks of people who doesn't even care about even reading the documentation of the filesystems they compare
OK, so ext3 is not the fastest filesystem on earth. But it has some default options which makes it suck even more than it usually do, and those options are *documented* in Documentation/filesystem/ext3.txt
* Ext3 does a sync() every 5 seconds. This is because ext3 developers are paranoid about your data and prefers to care about your data than win on benchmarks. Syncing every 5 seconds ensures you don't lose more than 5 seconds of work but it hurts on benchmarks. Other filesystems don't do it, if you are doing a FAIR comparison override the default with the "commit" mount option
* ext3's default journaling mode is slower than those from XFS, JFS or reiserfs, because it's safer. When ext3 is going to write some metadata to the journal, it takes care of writting to the disk the data associated to that metadata. XFS and JFS journaling modes do *not* care about this, neither they should, journaling was designed to keep filesystem integrity intact, not data, ext3 does it as an "extra", and it's slower because of that. But if you want to do a fair comparison, you should use the "data=writeback" mount option, which makes ext3 behave like xfs and jfs WRT to journaling. Reiserfs default journaling mode is like XFS/JFS, but you can make it behave like the ext3 default option with "data=ordered"
ext3 is not going to beat the other by using those mount options, but it won't suck so much, and the comparison will be more fair. And remember: ext3 tradeoffs data integrity for speed. There's nothing wrong with XFS and JFS, but _I_ use ext3.
He's right, to some point. When unix was created (linux is just a unix clone, bsd a unix derivate, windows is just a "unix sucks so we're going to reimplement the same ideas in a different way") there was no internet.
And then, internet was created. And TCP/IP was created on top of the current unix design, but nobody wasted time redesigning unix to make it not suck with networks.
The guys at the labs where unix was created (bell labs) realized that unix sucked. It sucked so much that they decided to redesign the whole OS, and they created Plan 9, a really beautiful OS where the system is really integrated in the network, but nobody cares about plan9 because, you know "unix rules" (notice the irony) and all that.
Take a look at one of the plan 9 papers to realize why unix sucks WRT to networks, and why current unix design can NOT really handle internet properly (regardless of that internet works thanks to big unix irons). It's time for the unix community to stop thinking that unix is the best operative system design you can get and start fixing it.
Those kernels are mostly written in C. They allow you to use C++ for drivers but they even "discourage" it in the documentation. Userspace is another matter.
As a linux kernel hacker said: "a language where '+' is allowed to do something that it's not really '+' is not a good choice for a kernel"
But I feel bad about adding more and more functions to KDE and leaving the rest of the system behind
Yes, FUSE rocks. But KDE was started in 1996 - FUSE was far from being "there"
It'd be a nice thing to se FUSE replacing KIO and gnome-vfs. SADLY, the freedesktop people have decided to start Yet Another Specification which aims to unify KIO and gnome-vfs. FUSE is the right thing to do - transparent to ALL apps even those compiled years ago - but no matter how much I explained to him, they think that the "Right Way" is creating a two-namespaces filesystem namespace. When you paste your URL from konqueror or nautilus to your terminal it won't work because you need to port cat to dvfs. (yes, we'd need a bsd FUSE equivalent - but such things are already happening)
It's sad that freedesktop people are accepting whatever project as long as it unifies "something" even if it doesn't have sense. There we go, creating a technically-crappy software base for the future of open source.
Slashdot Mirror
I didn't even knew who Steven Gibson was before this post. Russinovich's site (sysinternals.com) is one of the sites you can't stop visiting if you're doing anything with windows, even NT programmers at Microsoft use it, and Microsoft talks about those programs in several support articles. Just because people has know him after he discovered and analized the sony rootkit doesn't means he has never had a "reputation" as an expert.
Laptops running powerful dual-core CPUs eating less than 30W means nothing to you? Dualcore desktops alone are the biggest change in the CPU world since people started having computers in their homes, IMHO.
They'd have a platform that would whomp all over everything currently in the marketplace
Well, Alpha was a high-end CPU designed for servers. Somehow, I doubt it could whomp the portable market? (which is a important part of the computer world these days)
So - suddenly Intel, whose CPUs are the most power-hungry of the whole CPU industry, has a secret to make CPUs that are 2-3x more efficient (200-300%) than the ones from their competitors from now to 15-20 years (switching an architecture is a long-term strategy, not a short-term one)?
It's just me who thinks this is just bullshit? There are low-power 970GX chips that eat less than 30W (maybe they're not as good as centrino, but they're OK, would you risk the long-term strategy of your company for 5-10W?). Stop telling that "G5 are too hot to be put in a laptop".
It's like the "IBM can't provide a 3 GHz CPU" - AMD has been beating Intel with CPUs at frequencies lower than 3 GHz (right now AMD hasn't CPUs at 3 ghz, just a few 2.8 ghz ones); only intel has needed to go beyond of 3GHz to keep competitive and Jobs asks IBM to provide him a 3ghz CPU when he knows that 2.8 Ghz opterons are beating the 3.8 ghz intel monsters? I guess it's much easier to convince apple zealots that the intel move is the "right thing" if you hide the real reason and invent false excuses, right?
It's not just Intel. Until recently, +95% of the x86 server market did belong to Intel - that means that mobo/chipset/other products makers have their product offerings oriented to Intel. AMD made a great chip, but moving the whole x86 server market from one CPU vendor to another is not something that will happen from one day to another - Intel can't influde that, lots of companies are just not switching to another CPU vendor just because they've had good sales in 1 year. There're no good server-oriented chipsets for AMD chips except the ones from AMD for example (please don't bother mentioning that company that makes graphics cards and started making chipsets yesterday); however you can find chipsets from serverworks or IBM for Intel CPUs. I'm not suprised Intel keeps selling more than AMD, AMD will have to _keep_ the lead a few years if they want to get support from the industry.
I don't know others, but I haven't hit a page that I can't use with firefox in months. I know there're banks (which I don't use) and other stuff, but I think it's fair to say that the number of pages which doesn't work in firefox are 1) unmaintained (and hence, not worth of visiting) 2) a rare exception
What I do care about is that webmasters can't write pages use standars because IE doesn't supports it. Pages not rendering correctly in firefox are sad, but it's also bad that firefox supports lots of things that will never used in websites in a two years timeframe.
"Apple zealots don't like thinking . . . "
Well, there's a perfectly inane comment. Try some thinking yourself.
Hey, that was a perfect definition. Zealots, by definition, _are_ stupid. I never said "all the apple users" or something like that...
Read between the lines Jobs spoke and you find that Apple actually was struggling to come up with convincing evidence that justifies the Intel switch. Nonetheless, Jobs used wording that implied massive (2-4X!) improvements in performance, which, surprise surprise, are clearly massive exaggerations and, in the context of comparing this year's Pentiums to last year's G5s, are actually covers for practical failure on their side. ...not to mention the "IBM hasn't been able to deliver a 3 GHz chip" crap. Only Intel needed to go beyond 3 GHz to remain competitive, AMD has been the performance/MHz performance/watt leader for more than a year, and they didn't need to make 3 Ghz chip to do that (right now I think that AMD has not even released a 3 Ghz chip, just a 2.8 one) In other words, CPus didn't need to have more than 3 Ghz to remain competitive, only the crappy presscot Intel chips, and Jobs (who has spent years telling people that MHz are not important) knows this very well. He knew that G5 didn't need to go beyond 3 ghz to perform well
Still he promised a 3 Ghz CPU *two*years*before* announcing the intel switch (june 2003). To me, it looks like the Intel switch was decided some years ago, and Jobs said that crap just to have Yet Another excuse to justify the Intel switch to apple zealots. Of course, it has been a succesful move - Apple zealots don't like thinking, and they won't stop thinking twice how stupid the "IBM hasn't been able to deliver a 3 ghz CPU" excuse is.
Speaking of EFI, linux supports it. I've heard lot of noise about windows support, but what about linux support, will linux (and BSDs) be able to run on those things? I'm not buying a machine that can't run open operative systems...
Repeat after me: Mac OS X is not a microkernel. Mac OS X is not a microkernel
Mac OS X is derived from a microkernel (Mach) - but it's far from being a microkernel. A microkernel is not supposed to have the whole TCP/IP stack in kernel space. A microkernel is not supposed to implement drivers in kernel space. A microkernel is not supposed to have the filesystem in kernel space. Microkernels were, in fact, invented to get these things out of kernel space and run as userspace, etc. Being a pure microkernel implied a performance penaly they were not willing to pay. By moving all those things to kernel space, Mac OS X broke the whole "idea" behind microkernels and stopped being a microkernel (which is not a bad thing: all the other OSes do the same).
NT was also derived from Mach BTW. I will never understand why Mac zealots spend countless hours saying how crap the NT kernel is when the fact is that they derive from the same idea. And just because they are derived from a microkernel doesn't mean the microkernel is who implements all the funtionality. Mac OS X and NT copy from mach the "design": processes implementing funtionality runing in userspace, the real kernel being a scheduler scheduling those processes, etc. That's the "framework", the real functionality (TCP/IP, drivers, FS, etc) is implemented on top of that. The difference is that NT implemented that funcionality looking at VMS, and Mac OS X/nextstep implemented it by using BSD code from FreeBSD. There're some exceptions i think (mac os x vfs and journaling layer is implemented by Mach i think) but the idea is that. And this is BTW the reason why Linus Torvalds just hates mac os x and NT design: If you're going to do a microkernel, do it, but if you are going to put all the drivers and tcp/ip stack in the kernel, then don't use a microkernel design base because you're using the wrong tool and you're overengineering, just use a monolithic kernel and modularize it as much as you can (or something like that)
Bullshit. What last.fm does precisely is the same, and I don't see why this is bad. Take a look at my profile if you want to "spy" me. Of course, this is itunes and itunes is "good" and apple is "cool", so some people just feel the need to spread FUD in the hope they "free" people from itunes or "show people the reality behind itunes" or some crap like that - I call that "the superman syndrome"
The WMF format was created for windows 3.1, they may not work at microsoft anymore.
And the WMF "vulnerability" was NOT a vulnerability. It was a feature (you could attach executable code in the WMF for some reason, yes). Of course it was created when there was no internet, but it was a feature and it might have been a useful feature at that time.
It's unusual because Microsoft ALWAYS releases patches on thursdays. People has been asked for years to release them as soon as they're ready, and this time they had to release them sooner, because there were too many risks. The WMF vulnerability has been indeed unusual.
Oh, but this is free software and people packages things in several formats. Windows software is bundled with a _single_ installer and you can't change it.
One of the advantages of windows is that it comes preinstalled and users don't need to partition their hard disks. Linux distros need to force users to create partitions because there's no room for linux by default. Users also need to create a partition of the don't have room for XP. Partitions is not something windows can avoid.
To start with, you've to install tons of apps that the operative systems don't includes itself. And due to that stupid microsoft rule that existed for years ("installer must be executables delivered by 3rd party apps") I've no way to automate the download and installation of those (yes, I know about msi, I also know MSIs can be slipped in the installation CD. I still find no way of installing AND automatic its update like apt-get update & upgrade does. And LOTS of installers are not using MSI still. Shame on you microsoft, for forcing people to create docens of different, incompatible, buggy, installers)
Just because they hired a Gaim developer doesn't means he works in gaim....
http://code.google.com/
Is it just me or does this sound like a ridiculously ugly approach to data integrity?
It must be just you. By sync'ing your data every 5 seconds they ensure that you can't lose more of 5 seconds of work. Other filesystems try to avoid syncing as hard as possible and you could lose 10, 20, 30 seconds of work...
it appears that ext2/3 wins every single test that matters to me: find dirs, files, untar, tar, copy tarballs(s), the kernel source tree (ext2/3 now outperforms reiser3
It still loses on others. benchmarking it properly can change things...
Distros like Fedora that mandate the initial install ONLY use Ext3 are being stupid
It's amazing that such commentaries are moderated interesting these days. So, uh, fedora developers are stupid and you're smarter than them?. Please take a look at this commentary to understand why such decisions aren't so simple. You can tune your car's engine and it'll be faster, right? But why not everybody tunes their engines?
Let me quote a ext3 paper: "The ext2 and ext3 filesystems on Linux are used by a very large number of users. This is due to its reputation of dependability, robustness, backwards and forwards compatibility, rather than that of being the state of the art in filesystem technology."
I'm *sick* of reading filesystem benchmarks of people who doesn't even care about even reading the documentation of the filesystems they compare
OK, so ext3 is not the fastest filesystem on earth. But it has some default options which makes it suck even more than it usually do, and those options are *documented* in Documentation/filesystem/ext3.txt
* Ext3 does a sync() every 5 seconds. This is because ext3 developers are paranoid about your data and prefers to care about your data than win on benchmarks. Syncing every 5 seconds ensures you don't lose more than 5 seconds of work but it hurts on benchmarks. Other filesystems don't do it, if you are doing a FAIR comparison override the default with the "commit" mount option
* ext3's default journaling mode is slower than those from XFS, JFS or reiserfs, because it's safer. When ext3 is going to write some metadata to the journal, it takes care of writting to the disk the data associated to that metadata. XFS and JFS journaling modes do *not* care about this, neither they should, journaling was designed to keep filesystem integrity intact, not data, ext3 does it as an "extra", and it's slower because of that. But if you want to do a fair comparison, you should use the "data=writeback" mount option, which makes ext3 behave like xfs and jfs WRT to journaling. Reiserfs default journaling mode is like XFS/JFS, but you can make it behave like the ext3 default option with "data=ordered"
ext3 is not going to beat the other by using those mount options, but it won't suck so much, and the comparison will be more fair. And remember: ext3 tradeoffs data integrity for speed. There's nothing wrong with XFS and JFS, but _I_ use ext3.
He's right, to some point. When unix was created (linux is just a unix clone, bsd a unix derivate, windows is just a "unix sucks so we're going to reimplement the same ideas in a different way") there was no internet.
And then, internet was created. And TCP/IP was created on top of the current unix design, but nobody wasted time redesigning unix to make it not suck with networks.
The guys at the labs where unix was created (bell labs) realized that unix sucked. It sucked so much that they decided to redesign the whole OS, and they created Plan 9, a really beautiful OS where the system is really integrated in the network, but nobody cares about plan9 because, you know "unix rules" (notice the irony) and all that.
Take a look at one of the plan 9 papers to realize why unix sucks WRT to networks, and why current unix design can NOT really handle internet properly (regardless of that internet works thanks to big unix irons). It's time for the unix community to stop thinking that unix is the best operative system design you can get and start fixing it.
Windows XP, NT, 9x
Those kernels are mostly written in C. They allow you to use C++ for drivers but they even "discourage" it in the documentation. Userspace is another matter.
As a linux kernel hacker said: "a language where '+' is allowed to do something that it's not really '+' is not a good choice for a kernel"
But I feel bad about adding more and more functions to KDE and leaving the rest of the system behind
Yes, FUSE rocks. But KDE was started in 1996 - FUSE was far from being "there"
It'd be a nice thing to se FUSE replacing KIO and gnome-vfs. SADLY, the freedesktop people have decided to start Yet Another Specification which aims to unify KIO and gnome-vfs. FUSE is the right thing to do - transparent to ALL apps even those compiled years ago - but no matter how much I explained to him, they think that the "Right Way" is creating a two-namespaces filesystem namespace. When you paste your URL from konqueror or nautilus to your terminal it won't work because you need to port cat to dvfs. (yes, we'd need a bsd FUSE equivalent - but such things are already happening)
It's sad that freedesktop people are accepting whatever project as long as it unifies "something" even if it doesn't have sense. There we go, creating a technically-crappy software base for the future of open source.