It's not impossible, it's just orders of magnitude harder and less efficient.
And if, as Raymond claims, a significant chunk of debugging is done that way, it's no wonder why Windows is buggy to such a ridiculous degree.
Debian and probably most if not all other distributions mandate all packages to be compiled with -g for this reason, and when a toolchain problem on hppa broke -g in a good deal of cases, that was the stone that caused hppa to get dropped from squeeze.
I'd like to invite you to drive the Tricity Beltway near Szadólki. If you have your windows closed, a well-hermetic car, an air refresher and have a cold, you may survive. That smell is so rancid that any concerns for the environment will go away the moment you take a whiff.
Of course, they could have placed that landfill away from housing and from the road that greets almost every visitor to Gdansk, sparing them such a grand welcome.
It is useless for any sysadmin/developer work, and for any longer work with text. But ordinary users rarely do those.
For them, letting them view YouTube or play some game without keyboard on a device that costs less due to that is a valid tradeoff. Not for us, though.
Because we suffer from some bozo's religion. If for you following it is more important that avoiding getting crippled and dying to a disease that makes the last 15 years of your life a hell for your family, it shouldn't stop me from having a portion of my tax money used in an attempt to keep me from that fate. And yeah, I had Alzheimer's on both sides of my family too.
To begin with 99% of commercial games don't even have a Linux version, so there's nothing to sell to Linux gamers.
The same could be said of Macs. Part of what made Steam viable on the Mac was Valve porting a number of their games over to the Mac. And they could do it again for Linux if they wanted to...
I guess it's not really about porting the games itself, it's about porting the DRM. Any form of DRM is fundamentally incompatible with an open OS -- since the user has full control over the hardware, and can tell the kernel to lie to the Steam client in any way he wants. And even if they add a kernel driver, it can be easily worked around.
On Windows and Mac, DRM, while still not viable in theory, works in practice since working around it is hard.
They do spend money (not theirs, taxpayers') to make it look they do care for the environment, which brings votes. They don't give a flying damn about doing something that actually works.
As soon as Apple releases an iPhone with a slide out QWERTY keyboard, I'm in.
Why would you bother? That's a completely different class of gear.
iPhone is a phone with a bunch of toys, N900 is a full sub-notebook with phone capabilities tacked on. The former can run just a few random "apps", the latter allows you to install a regular OS with all of its functionality.
The keyboard is one of significant advantages of N900, but definitely not the main one. For one, the research done in this article would be flat out impossible on iPhone due to its closed nature.
"The only area"... that's in a pinch, anything that ends in "byte".
And for calculations, what makes powers of two worse than powers of ten? With the subject matter at hand, it has significant advantage. And for drive sizes, have you ever looked at filesystem internals? With sector size, page size, all internal divisions in filesystems I know being powers of two, that's vital -- not "completely meaningless and useless".
Not including PEOPLE, I understand and agree with that. But buildings that are visible from public streets are a public matter, and anyone trying to block the photos is infringing on the public's rights.
My sister's ex' sister is a drug rep, and she was describing exactly what the GP says, except for "coke and hookers", but with coke being nearly unheard of around here, that's pretty consistent.
And just go visit any doctor, or if that's your family, just see what they're doing. Having all lapels in their coats, all pens and all pads bear the logo of a pharma company, then the walls covered in such posters, quite suggest there might be some way too tight relations...
Uhm, Compiz useless? Have you tried it on a computer with a semi-decent graphics card but a lousy CPU? Even on a modern one, without Compiz you often have to wait until windows draw themselves whenever you switch desktops -- with Compiz, it's instanteous. The worst hardware I've seen it on a P4-era Celeron with an nVidia 5600 -- and the speed gains from Compiz there are just insane.
If it takes 100% CPU, this means you don't have 3D acceleration in working order -- or that you tried to turn on every single gimmick at the same time. The whole point of Compiz is to use the GPU not CPU. Once a window is drawn, it is stored in the texture memory, which means it can be displayed without the main CPU's help. This happens to be what makes switching desktops that fast.
Gnome vs KDE is another story... and with all of Gnome's downsides, I'd say KDE currently can't hold a candle to it; it's an emacs-vs-vi thing, though.
Having tried multitouch, it's useless in the long term. It is a nice gimmick to show in an advertisement, but for using it for longer than 15 minutes at a time, it's not a good idea -- you'll hand will get sore in no time.
Even for mobile devices, there is simply no better thing than the good old keyboard. If you try the on-screen touch thingy on an iPad or most Androids, it may be enough for typing a single line of text. On an N900 with a proper physical keyboard, you're in good shape after several hours of typing. And since you can't have that many distinct gestures, traditional keyboard shortcuts are so much better.
What "Internet Explorer" are you talking about? It has already been renamed to "Windows Internet Explorer", so the contest which name will last longer is already over.
Hey, that cover is not enough to stop a crafty 3 years old, not just a curious temp.
As a toddler, at mom's work (kindergarten was only for families of members of the Party...), I turned off a row of disk drives -- of the washing machine sized kind, standing far away from the actual computers. They had such molly guards and had their power switches far up, but it was a matter of standing on my toes and lifting the flap. And I repeated that twice on other drives before getting stopped. And funnily enough, they banned kids from that place only when I did it again on another day:p
iPhones and similar stuff are toys for mainstream users. n900 is for serious sysadmin (or even programmer) work. Of course the second group tends to get no sex.
The physical keyboard makes it a subnotebook with superior portability and networking capabilities. It has its flaws -- having to press Fn to get numbers makes typing, say, IP addresses a bitch, and 256MB main RAM cripples compilation speed even if the CPU is decent, but in the two weeks I have one, I spent productively a lot of time that would be otherwise wasted, in places I wouldn't take a laptop to.
These just use the good old Unix memory management -- if you can coordinate between multiple VMs, things get a whole lot easier. The problem with VMs with separate kernels (Xen, VirtualBox, VMWare, etc) is that they have no way of knowing a given page mmaps the same block on the disk.
The technique described in the article is a hack that works only if all processes are started before you clone the VMs and nothing else happens later. Vserver does it strictly better -- if multiple VMs use the same file on the disk, it will use the memory exactly once, no matter when it was read.
Most of modern filesystems don't put the new data into the old place. This is most prominent on JFFS (which is mostly the entire reason for it), then, in a decreasing order: btrfs, reiserfs, jfs, ext[34]. And on old filesystems on flash, you'll often have an underlying layer that does wear-levelling. Also, if there's any copy-on-write, tail packing, snapshots, etc, involved, shred will most likely be defeated as well.
It's not impossible, it's just orders of magnitude harder and less efficient.
And if, as Raymond claims, a significant chunk of debugging is done that way, it's no wonder why Windows is buggy to such a ridiculous degree.
Debian and probably most if not all other distributions mandate all packages to be compiled with -g for this reason, and when a toolchain problem on hppa broke -g in a good deal of cases, that was the stone that caused hppa to get dropped from squeeze.
I'd like to invite you to drive the Tricity Beltway near Szadólki. If you have your windows closed, a well-hermetic car, an air refresher and have a cold, you may survive. That smell is so rancid that any concerns for the environment will go away the moment you take a whiff.
Of course, they could have placed that landfill away from housing and from the road that greets almost every visitor to Gdansk, sparing them such a grand welcome.
OT: screw you /. for that great Unicode support.
It is useless for any sysadmin/developer work, and for any longer work with text. But ordinary users rarely do those.
For them, letting them view YouTube or play some game without keyboard on a device that costs less due to that is a valid tradeoff. Not for us, though.
And why does that override moral concerns?
Because we suffer from some bozo's religion. If for you following it is more important that avoiding getting crippled and dying to a disease that makes the last 15 years of your life a hell for your family, it shouldn't stop me from having a portion of my tax money used in an attempt to keep me from that fate. And yeah, I had Alzheimer's on both sides of my family too.
To begin with 99% of commercial games don't even have a Linux version, so there's nothing to sell to Linux gamers.
The same could be said of Macs. Part of what made Steam viable on the Mac was Valve porting a number of their games over to the Mac. And they could do it again for Linux if they wanted to...
I guess it's not really about porting the games itself, it's about porting the DRM. Any form of DRM is fundamentally incompatible with an open OS -- since the user has full control over the hardware, and can tell the kernel to lie to the Steam client in any way he wants. And even if they add a kernel driver, it can be easily worked around.
On Windows and Mac, DRM, while still not viable in theory, works in practice since working around it is hard.
They do spend money (not theirs, taxpayers') to make it look they do care for the environment, which brings votes. They don't give a flying damn about doing something that actually works.
You can say exactly the same about laptops, and somehow quite a lot of people use them.
As soon as Apple releases an iPhone with a slide out QWERTY keyboard, I'm in.
Why would you bother? That's a completely different class of gear.
iPhone is a phone with a bunch of toys, N900 is a full sub-notebook with phone capabilities tacked on.
The former can run just a few random "apps", the latter allows you to install a regular OS with all of its functionality.
The keyboard is one of significant advantages of N900, but definitely not the main one.
For one, the research done in this article would be flat out impossible on iPhone due to its closed nature.
It's not an alternate account, it's my only one.
I've used that nick since around 1995, way before that good-for-nothing committee that decided to redefine pi^Hkilobyte was formed.
"The only area"... that's in a pinch, anything that ends in "byte".
And for calculations, what makes powers of two worse than powers of ten? With the subject matter at hand, it has significant advantage. And for drive sizes, have you ever looked at filesystem internals? With sector size, page size, all internal divisions in filesystems I know being powers of two, that's vital -- not "completely meaningless and useless".
A Terabyte is what, 1000GB?
No, 1024GB. It's only drive makers and a committee that try to redefine that.
Don't they have 30GB caps in Japan?
Per day, of course.
Not including PEOPLE, I understand and agree with that. But buildings that are visible from public streets are a public matter, and anyone trying to block the photos is infringing on the public's rights.
These days, it will be a smartphone.
My sister's ex' sister is a drug rep, and she was describing exactly what the GP says, except for "coke and hookers", but with coke being nearly unheard of around here, that's pretty consistent.
And just go visit any doctor, or if that's your family, just see what they're doing. Having all lapels in their coats, all pens and all pads bear the logo of a pharma company, then the walls covered in such posters, quite suggest there might be some way too tight relations...
redefine it.
Uhm, Compiz useless? Have you tried it on a computer with a semi-decent graphics card but a lousy CPU? Even on a modern one, without Compiz you often have to wait until windows draw themselves whenever you switch desktops -- with Compiz, it's instanteous. The worst hardware I've seen it on a P4-era Celeron with an nVidia 5600 -- and the speed gains from Compiz there are just insane.
If it takes 100% CPU, this means you don't have 3D acceleration in working order -- or that you tried to turn on every single gimmick at the same time. The whole point of Compiz is to use the GPU not CPU. Once a window is drawn, it is stored in the texture memory, which means it can be displayed without the main CPU's help. This happens to be what makes switching desktops that fast.
Gnome vs KDE is another story... and with all of Gnome's downsides, I'd say KDE currently can't hold a candle to it; it's an emacs-vs-vi thing, though.
Having tried multitouch, it's useless in the long term. It is a nice gimmick to show in an advertisement, but for using it for longer than 15 minutes at a time, it's not a good idea -- you'll hand will get sore in no time.
Even for mobile devices, there is simply no better thing than the good old keyboard. If you try the on-screen touch thingy on an iPad or most Androids, it may be enough for typing a single line of text. On an N900 with a proper physical keyboard, you're in good shape after several hours of typing. And since you can't have that many distinct gestures, traditional keyboard shortcuts are so much better.
What "Internet Explorer" are you talking about? It has already been renamed to "Windows Internet Explorer", so the contest which name will last longer is already over.
And, also: die, WIE, die!
Hey, that cover is not enough to stop a crafty 3 years old, not just a curious temp.
As a toddler, at mom's work (kindergarten was only for families of members of the Party...), I turned off a row of disk drives -- of the washing machine sized kind, standing far away from the actual computers. They had such molly guards and had their power switches far up, but it was a matter of standing on my toes and lifting the flap. And I repeated that twice on other drives before getting stopped. And funnily enough, they banned kids from that place only when I did it again on another day :p
I spent productively a lot of time that would be otherwise wasted having sex
That would kind of distract the driver, or spook the passers-by, depending on what time you're talking about :p
iPhones and similar stuff are toys for mainstream users. n900 is for serious sysadmin (or even programmer) work. Of course the second group tends to get no sex.
The physical keyboard makes it a subnotebook with superior portability and networking capabilities. It has its flaws -- having to press Fn to get numbers makes typing, say, IP addresses a bitch, and 256MB main RAM cripples compilation speed even if the CPU is decent, but in the two weeks I have one, I spent productively a lot of time that would be otherwise wasted, in places I wouldn't take a laptop to.
Or vserver. Or BSD jails.
These just use the good old Unix memory management -- if you can coordinate between multiple VMs, things get a whole lot easier. The problem with VMs with separate kernels (Xen, VirtualBox, VMWare, etc) is that they have no way of knowing a given page mmaps the same block on the disk.
The technique described in the article is a hack that works only if all processes are started before you clone the VMs and nothing else happens later. Vserver does it strictly better -- if multiple VMs use the same file on the disk, it will use the memory exactly once, no matter when it was read.
Except that the speed he stated was 1.0000000033c, and that means he went back in time, at least for some observers.
Most of modern filesystems don't put the new data into the old place. This is most prominent on JFFS (which is mostly the entire reason for it), then, in a decreasing order: btrfs, reiserfs, jfs, ext[34]. And on old filesystems on flash, you'll often have an underlying layer that does wear-levelling. Also, if there's any copy-on-write, tail packing, snapshots, etc, involved, shred will most likely be defeated as well.