That's great an all if you want the file in the itunes format. Converting the AAC to something else however certainly entails loss from that which will without a doubt end up lower quality than the original. I want to be able to convert the original to the format of my choice, and play it on the player of my choice, in the OS of my choice. WITHOUT someone having to figure out how to crack the DRM.
Saying Knoppix AND anything that's been mentioned is silly. Parted, dd, and tar are all PART of knoppix.
I agree with what your really trying to say however in regards to tar but NOT dd. DD is quite useful in a school or corporate environment because with a proper sector size set DD is often actually faster than tar (hard to believe I know, but DD reads from one sector next physically inline sector instead of jumping around, the more fragmented the filesystem, the more of an advantage that DD has... on the other hand tar also defrags said filesystem). In these environments the disks and everything else are typically identical.
Where DD shines most however is for HDD imaging for data recovery purposes. Imaging the original drive to a new one with DD will allow you to attempt to reconstruct the data on your copy rather than risking doing further damage to the original. With DD you don't have to mount the original or perform any writes whatsoever to it and you copy every bit whether the filesystem thinks it's important/damaged or not. This can be a lifesaver.
Re:There are many limitations to Sysprep:
on
Knoppix Tips and Tricks
·
· Score: 2, Informative
Just to debunk (not by any means denying that's the official microsoft voice on the issue), as usual that is mostly microsoft trying to scare people away.
Mainly this, Sysprep works perfectly well on OEM versions AND upgrades.
Basically all sysprep REALLY does is, change the pc name and SID + knock out motherboard and ide drivers.
It's kind of like performing the first half of a win98 install, then ghosting that to different systems, you will get clean hardware detection every time and simply cut the install time down (since copying everything with ghost or knoppix at that point takes about 2 min where the installer takes 20).
Activation issues are not an issue.
So most of those things work (especially the big ones), just don't expect microsoft to help you with them. Have you ever called microsoft anyway? and if so, did ever make the same mistake (of calling) again?
yes, that is why you partition the destination drive, move contents via tar to partition on the new drive. Tada, you have not only cloned the drive, you have only copied all the data AND defragged simultaneously (something ghost does not do, you'd think it does but if you check the drive after, you'll find you were wrong).
If your concerned about NTFS write problems, don't be, they are mostly a myth and have been for some time. NTFS write is long past the stage of being "experimental" just out of pure stubborness, besides that it's not like you've lost anything but 10mins if the write fails and you can play with it where that's not a killer until you've done it enough to know it's not an issue.
If your concerned about needing lilo or grub because it's a linux drive, don't be. Either one can be loaded into the MBR of the new drive allowing it to be bootable. If it's windows, definately don't be concerned, losing the disk signiture isn't going to hurt anything and 9x versions have a blank mbr (the install blanks the MBR to ensure it's contents won't interfere, that's why it kills lilo/grub, not because it actually uses the MBR for something). XP and 2K bootloaders don't load from the mbr.
If you are going to an identical disk (and you often are when imaging) you can simply dd the mbr and partition table and then tar the rest.
It's still labeled experimental, however in my experience it's actually pretty stable nowdays. Haven't had any corruptions using it in the last year or so. Write support that is, read has been rock solid for a LONG time.
The experimental label isn't much of a big deal for copying drives anyway. After all, if your copying a drive, your copying it from somewhere, so if things go wrong you've lost 10mins but you haven't lost any data. That was my mindset when I first started testing it out with a system here and there. After awhile I figured out it was reliable enough to take to the other side of the country to do 200 systems.
Who said anything about either of those systems being 16bit DOS? Hell you could run 32bit applications on windows for workgroups. However that is really irrelevant in terms of whether or not the GUI was part of the OS, or the CLI was part of the OS.
Go find yourself a win95 or win98 system, open up msdos.sys in a text editor, change the bootgui option from 1 to 0. Then you will see whether the OS is CLI or GUI. Now in 98 if you open a command prompt WITHIN windows you get a dos emulator, running within a gui, running on top of dos. Kinda of pathetic isn't it? 90% of your old dos programs will run just fine if you turn of the gui and run them there. They don't work as well in the emulation on top of gui on top of cli.
Actually one restricts in another manner than the other. One requires acknowledgements. One requires open code in turn.
Whether you term it restrict or protect is really a glass half full or glass half empty.
Basically it's a matter of whether or not your a programmer. If your not a programmer (or a company who produces programs or has the ability to hire programmers) it makes no difference whatsoever. If you are a programmer (or company who produces program or has the ability to hire programmers) then it depends on whether or not you simply want a free lunch from open source so you can profit from the blood and sweat of others at no expense to yourself. The "restriction" you speak of is a benefit to absolutely EVERY other party since it means all of the benefits of open source will still apply to something which is based on open source code.
Now, if your a closed source company simply wanting the fruits of free slave labor. Then you call it a restriction rather than a protection.
Open source code after all isn't primarily about getting something for nothing, it's about being able to modify and improve something having your work be part of a greater whole. It's about knowing that when it comes down to it you don't have to rely on anyone else because you have the code and can make the changes you need be it features, bug fixes, behavior in this lil window here, etc. Even a non-programmer can hire a programmer to do these things. A 30 pc business can hire a coder to add a completely new file format to openoffice for less than it would code to purchase licenses of MS office for those 30 pc's ONCE let alone every couple years.
Linux does have support for writing to NTFS, it's been labeled experimental since it first came out. Hell it deserved it back then, corrupted everything in sight. As of about a year ago that magically disappeared (although the label has not, still experimental) and I've been using it since on a daily basis without a single corruption.
Hopefully this will get modded up or some such so I don't have to repeat it anymore.
You can if your laptops come pre-loaded with linux. May you don't have the market share potential out of the gate that windows companies have. But you'll have very little competition and literally millions of interested customers who won't mind paying a couple bucks more for something that works right. Of course the key is to get slashdotted, there could be no cheaper advertising than clicking submit.
True enough, DD != Ghost, but not what he claimed, he claimed that linux on a cd will supplant ghost and that is something different altogether.
Now your not talking about ghost, your talking about a number of tools.
mount partd mkfs kernel support for more filesystems than ghost will ever dream of. tar dd cp mkswap lilo/grub
Between these utilities you can do pretty much everything ghost can and much much more. A knoppix cd (generally I use a customized one to take out the gui fluff) gives FAR more flexibility than any other software tool.
Considering that according to EU estimates there over 2million open source developers alone (and you can bet every one of them knows how to set the string in their browser) and it doesn't take anywhere NEAR developer level knowledge to accomplish the task (you go into preferences in firebird for instance click the get plugins link and it takes you to a page which has a user agent switcher applet right there, at which point it's a menu option to switch between common browsers). I suspect you'll find that more than 1% of the net is doing this.
Yes, all of the old updates are available for redhat linux 5.2. The source code is also available if you wish to back port patches yourself (you can hire a poorly trained monkey to do this).
There was a pretty graphical interface called windows for workgroups for old dos, there was a pretty graphical interface on top of dos called win95 after that. and there was a pretty graphical interface called win98 after that. But they are all good old dos underneath.
That's great, if your having a problem per say. I don't think there are really many people out there who have called microsoft twice. On the other hand I don't think there is anyone out there who has used a Microsoft product and only had 1 problem they weren't able to figure out themselves. Microsoft support is garbage anyway, everyone goes to third parties.
The only thing people go to microsoft for is Updates and security patches. Those require the source.
"Haven't you ever seen a forum killed, or nearly so, from being overrun by the willfully ignorant?"
*looks around slashdot* Is this a trick question?
"I am on a FreeBSD list, and I'll tell you, there is no one like that there. I've never seen one, ever. Anybody who can get FreeBSD installed has at least two clues to rub together, and they will grow from there. Linux was once that way, and I don't see it becoming less that way as a good thing. Popularity has a price, and it's steep."
I see your logic now, and understand what you mean. But your logic is flawed and wishful thinking. The degradation of signal to noise is inevitable. If all those who are in the know move to BSD, BSD will slowly evolve into a feature rich platform. Slowly the not entirely ignorant but not thinking they need to be less so will migrate to it to be l33t. Your signal to noise will start to degrade, their suggestions and minor developments will start to nudge things in another direction. BSD would start to gain popularity, with popularity means more developers and that means more people to try to convince that things should be uneccesarily difficult in order to keep those who don't know what they are doing from the userbase.
Basically, it's inevitable, moving to BSD simply means recreating the wheel all over again. In the end it will be the same. Personally I'm going to stick with linux, I can't stop it from growing, eventually I may help new users less because of the signal to noise ratio. But they won't need as much help either as things become easier and more commercial support options become available. So I'll invest my time and input into linux and seeing that it remains flexible and solid.
Every man to his own I suppose. I simply don't see what's to gain by moving to BSD and starting the process to the same end all over again, and without the philosophical benefits of the GPL behind it. What follows when BSD gets popular? Hurd? Then when we've reinvented 10,000 wheels 3 times for each system what follows? Besides that some wheels are simply ported and not reinvented, so the decline of BSD would likely be faster than that of linux. And hurd"?" after that more rapid yet.
Seems better to switch forums than Operating system's to me.
"As for Linux on the desktop, the only argument I really have for it now is that I'd rather see anything other than MS on the desktop, and after Microsoft is will and thoroughly crushed, I can retreat into FreeBSD. It has an installer strong enough to keep away the computer-stupid. Already, the clue level in Linux has fallen so far due to the influx of newbies who are ignorant, want to stay that way, and don't want to learn the Unix Way, that I'm contemplating to jumping ship to BSD this year anyway."
I fail to see your logic. There are more ignorant people using linux and so you want to move to BSD? Why? More ignorant people hardly means less competant ones. If you have 20 competant users and add 1000 incompetant ones, you STILL have 20 competant users, the number hasn't magically reduced. Although the odds in reality are that if change that to just add 1000 users you will get another 10 who are competant and now have 30 competant users.
Yes the sys admin downloads the software, tests it to ensure stability and then adds it to his system image. Then sends his army of monkey's out to install it on your workstations.
It's not like you'll really need more than one or two tools that aren't already provided and it's not like those will ever have to be installed again.
If this actually results in a significant amount of time on an ongoing basis then obviously it's more productive for you to use the tools you have than to save 2 seconds by using a different one. Using one you "prefer" but that doesn't actually save time is a waste of course.
I'm an admin in one as a matter of fact. There are only two job roles that I can think of that require the ability to install software. Sysadmin and developer.
Most companies don't have developers for starters. And although developers need to install software, they DON'T need to install it on the corporate network. You give you them a few computers, a switch, a hub, and 2pc's worth of spare parts, and an annual budget of about $100 for it. If they break it and can't fix it, it's their own problem.
After all, it's certainly that way for the sysadmin's own test network.
This has happened before. What happens is that the GPL'd copies are still GPL'd and therefore not in violation of any license. Once you distribute something under the GPL, it is GPL'd period.
There is nothing which stops you from no longer providing licenses under the GPL (in which your case your development from that point for wouldn't be open to all), nor is there anything preventing someone else from taking what you had released up to that point and forking it, providing updates to it as well since the GPL license you distributed to them continues. The GPL can't be revoked short of a violation of it's terms.
Slashdot Mirror
That's great an all if you want the file in the itunes format. Converting the AAC to something else however certainly entails loss from that which will without a doubt end up lower quality than the original. I want to be able to convert the original to the format of my choice, and play it on the player of my choice, in the OS of my choice. WITHOUT someone having to figure out how to crack the DRM.
WARNING THIS IS A TROLL, SORT OF ;)
;)
If I used debian I wouldn't want to ever install again either
Saying Knoppix AND anything that's been mentioned is silly. Parted, dd, and tar are all PART of knoppix.
I agree with what your really trying to say however in regards to tar but NOT dd. DD is quite useful in a school or corporate environment because with a proper sector size set DD is often actually faster than tar (hard to believe I know, but DD reads from one sector next physically inline sector instead of jumping around, the more fragmented the filesystem, the more of an advantage that DD has... on the other hand tar also defrags said filesystem). In these environments the disks and everything else are typically identical.
Where DD shines most however is for HDD imaging for data recovery purposes. Imaging the original drive to a new one with DD will allow you to attempt to reconstruct the data on your copy rather than risking doing further damage to the original. With DD you don't have to mount the original or perform any writes whatsoever to it and you copy every bit whether the filesystem thinks it's important/damaged or not. This can be a lifesaver.
Just to debunk (not by any means denying that's the official microsoft voice on the issue), as usual that is mostly microsoft trying to scare people away.
Mainly this, Sysprep works perfectly well on OEM versions AND upgrades.
Basically all sysprep REALLY does is, change the pc name and SID + knock out motherboard and ide drivers.
It's kind of like performing the first half of a win98 install, then ghosting that to different systems, you will get clean hardware detection every time and simply cut the install time down (since copying everything with ghost or knoppix at that point takes about 2 min where the installer takes 20).
Activation issues are not an issue.
So most of those things work (especially the big ones), just don't expect microsoft to help you with them. Have you ever called microsoft anyway? and if so, did ever make the same mistake (of calling) again?
yes, that is why you partition the destination drive, move contents via tar to partition on the new drive. Tada, you have not only cloned the drive, you have only copied all the data AND defragged simultaneously (something ghost does not do, you'd think it does but if you check the drive after, you'll find you were wrong).
If your concerned about NTFS write problems, don't be, they are mostly a myth and have been for some time. NTFS write is long past the stage of being "experimental" just out of pure stubborness, besides that it's not like you've lost anything but 10mins if the write fails and you can play with it where that's not a killer until you've done it enough to know it's not an issue.
If your concerned about needing lilo or grub because it's a linux drive, don't be. Either one can be loaded into the MBR of the new drive allowing it to be bootable. If it's windows, definately don't be concerned, losing the disk signiture isn't going to hurt anything and 9x versions have a blank mbr (the install blanks the MBR to ensure it's contents won't interfere, that's why it kills lilo/grub, not because it actually uses the MBR for something). XP and 2K bootloaders don't load from the mbr.
If you are going to an identical disk (and you often are when imaging) you can simply dd the mbr and partition table and then tar the rest.
It's still labeled experimental, however in my experience it's actually pretty stable nowdays. Haven't had any corruptions using it in the last year or so. Write support that is, read has been rock solid for a LONG time.
The experimental label isn't much of a big deal for copying drives anyway. After all, if your copying a drive, your copying it from somewhere, so if things go wrong you've lost 10mins but you haven't lost any data. That was my mindset when I first started testing it out with a system here and there. After awhile I figured out it was reliable enough to take to the other side of the country to do 200 systems.
Who said anything about either of those systems being 16bit DOS? Hell you could run 32bit applications on windows for workgroups. However that is really irrelevant in terms of whether or not the GUI was part of the OS, or the CLI was part of the OS.
Go find yourself a win95 or win98 system, open up msdos.sys in a text editor, change the bootgui option from 1 to 0. Then you will see whether the OS is CLI or GUI. Now in 98 if you open a command prompt WITHIN windows you get a dos emulator, running within a gui, running on top of dos. Kinda of pathetic isn't it? 90% of your old dos programs will run just fine if you turn of the gui and run them there. They don't work as well in the emulation on top of gui on top of cli.
Actually one restricts in another manner than the other. One requires acknowledgements. One requires open code in turn.
Whether you term it restrict or protect is really a glass half full or glass half empty.
Basically it's a matter of whether or not your a programmer. If your not a programmer (or a company who produces programs or has the ability to hire programmers) it makes no difference whatsoever. If you are a programmer (or company who produces program or has the ability to hire programmers) then it depends on whether or not you simply want a free lunch from open source so you can profit from the blood and sweat of others at no expense to yourself. The "restriction" you speak of is a benefit to absolutely EVERY other party since it means all of the benefits of open source will still apply to something which is based on open source code.
Now, if your a closed source company simply wanting the fruits of free slave labor. Then you call it a restriction rather than a protection.
Open source code after all isn't primarily about getting something for nothing, it's about being able to modify and improve something having your work be part of a greater whole. It's about knowing that when it comes down to it you don't have to rely on anyone else because you have the code and can make the changes you need be it features, bug fixes, behavior in this lil window here, etc. Even a non-programmer can hire a programmer to do these things. A 30 pc business can hire a coder to add a completely new file format to openoffice for less than it would code to purchase licenses of MS office for those 30 pc's ONCE let alone every couple years.
Linux does have support for writing to NTFS, it's been labeled experimental since it first came out. Hell it deserved it back then, corrupted everything in sight. As of about a year ago that magically disappeared (although the label has not, still experimental) and I've been using it since on a daily basis without a single corruption.
Hopefully this will get modded up or some such so I don't have to repeat it anymore.
yup, why I imagine you can even view 10% of the web that way!!! Although it could be argued that it's the 10% that counts ;)
Actually you can't if you make a ghost boot cd. However you can with a ghost boot floppy, just like you can with a linux boot floppy.
no...
;P
merely equally skilled
why? Both are on one disk and equally easy to use for someone who already knows how to use the tool.
If the choice for the task is the linux distro vs ghost wouldn't the proper comparison be between those rather than ghost vs an individual util?
You can if your laptops come pre-loaded with linux. May you don't have the market share potential out of the gate that windows companies have. But you'll have very little competition and literally millions of interested customers who won't mind paying a couple bucks more for something that works right. Of course the key is to get slashdotted, there could be no cheaper advertising than clicking submit.
True enough, DD != Ghost, but not what he claimed, he claimed that linux on a cd will supplant ghost and that is something different altogether.
Now your not talking about ghost, your talking about a number of tools.
mount
partd
mkfs
kernel support for more filesystems than ghost will ever dream of.
tar
dd
cp
mkswap
lilo/grub
Between these utilities you can do pretty much everything ghost can and much much more. A knoppix cd (generally I use a customized one to take out the gui fluff) gives FAR more flexibility than any other software tool.
Considering that according to EU estimates there over 2million open source developers alone (and you can bet every one of them knows how to set the string in their browser) and it doesn't take anywhere NEAR developer level knowledge to accomplish the task (you go into preferences in firebird for instance click the get plugins link and it takes you to a page which has a user agent switcher applet right there, at which point it's a menu option to switch between common browsers). I suspect you'll find that more than 1% of the net is doing this.
I'd argue that there are more people out there using redhat/fedora download versions at present then will EVER use redhat enterprise.
Yes, all of the old updates are available for redhat linux 5.2. The source code is also available if you wish to back port patches yourself (you can hire a poorly trained monkey to do this).
There was a pretty graphical interface called windows for workgroups for old dos, there was a pretty graphical interface on top of dos called win95 after that. and there was a pretty graphical interface called win98 after that. But they are all good old dos underneath.
That's great, if your having a problem per say. I don't think there are really many people out there who have called microsoft twice. On the other hand I don't think there is anyone out there who has used a Microsoft product and only had 1 problem they weren't able to figure out themselves. Microsoft support is garbage anyway, everyone goes to third parties.
The only thing people go to microsoft for is Updates and security patches. Those require the source.
"Haven't you ever seen a forum killed, or nearly so, from being overrun by the willfully ignorant?"
*looks around slashdot* Is this a trick question?
"I am on a FreeBSD list, and I'll tell you, there is no one like that there. I've never seen one, ever. Anybody who can get FreeBSD installed has at least two clues to rub together, and they will grow from there. Linux was once that way, and I don't see it becoming less that way as a good thing. Popularity has a price, and it's steep."
I see your logic now, and understand what you mean. But your logic is flawed and wishful thinking. The degradation of signal to noise is inevitable. If all those who are in the know move to BSD, BSD will slowly evolve into a feature rich platform. Slowly the not entirely ignorant but not thinking they need to be less so will migrate to it to be l33t. Your signal to noise will start to degrade, their suggestions and minor developments will start to nudge things in another direction. BSD would start to gain popularity, with popularity means more developers and that means more people to try to convince that things should be uneccesarily difficult in order to keep those who don't know what they are doing from the userbase.
Basically, it's inevitable, moving to BSD simply means recreating the wheel all over again. In the end it will be the same. Personally I'm going to stick with linux, I can't stop it from growing, eventually I may help new users less because of the signal to noise ratio. But they won't need as much help either as things become easier and more commercial support options become available. So I'll invest my time and input into linux and seeing that it remains flexible and solid.
Every man to his own I suppose. I simply don't see what's to gain by moving to BSD and starting the process to the same end all over again, and without the philosophical benefits of the GPL behind it. What follows when BSD gets popular? Hurd? Then when we've reinvented 10,000 wheels 3 times for each system what follows? Besides that some wheels are simply ported and not reinvented, so the decline of BSD would likely be faster than that of linux. And hurd"?" after that more rapid yet.
Seems better to switch forums than Operating system's to me.
"As for Linux on the desktop, the only argument I really have for it now is that I'd rather see anything other than MS on the desktop, and after Microsoft is will and thoroughly crushed, I can retreat into FreeBSD. It has an installer strong enough to keep away the computer-stupid. Already, the clue level in Linux has fallen so far due to the influx of newbies who are ignorant, want to stay that way, and don't want to learn the Unix Way, that I'm contemplating to jumping ship to BSD this year anyway."
I fail to see your logic. There are more ignorant people using linux and so you want to move to BSD? Why? More ignorant people hardly means less competant ones. If you have 20 competant users and add 1000 incompetant ones, you STILL have 20 competant users, the number hasn't magically reduced. Although the odds in reality are that if change that to just add 1000 users you will get another 10 who are competant and now have 30 competant users.
Yes the sys admin downloads the software, tests it to ensure stability and then adds it to his system image. Then sends his army of monkey's out to install it on your workstations.
It's not like you'll really need more than one or two tools that aren't already provided and it's not like those will ever have to be installed again.
If this actually results in a significant amount of time on an ongoing basis then obviously it's more productive for you to use the tools you have than to save 2 seconds by using a different one. Using one you "prefer" but that doesn't actually save time is a waste of course.
I'm an admin in one as a matter of fact. There are only two job roles that I can think of that require the ability to install software. Sysadmin and developer.
Most companies don't have developers for starters. And although developers need to install software, they DON'T need to install it on the corporate network. You give you them a few computers, a switch, a hub, and 2pc's worth of spare parts, and an annual budget of about $100 for it. If they break it and can't fix it, it's their own problem.
After all, it's certainly that way for the sysadmin's own test network.
This has happened before. What happens is that the GPL'd copies are still GPL'd and therefore not in violation of any license. Once you distribute something under the GPL, it is GPL'd period.
There is nothing which stops you from no longer providing licenses under the GPL (in which your case your development from that point for wouldn't be open to all), nor is there anything preventing someone else from taking what you had released up to that point and forking it, providing updates to it as well since the GPL license you distributed to them continues. The GPL can't be revoked short of a violation of it's terms.