Edgy and Feisty are only supported _by Ubuntu_ for 18 months. There's no reason Dell can't provide, or pay another 3rd party to provide, or even pay Ubuntu to provide, longer term support for the version of Ubuntu they ship. That's one of the beauties of Free Software - anyone can provide support for any bit of software, because anyone can go in there and fix bugs, even after upstream has stopped supporting that version.
Um....but these people have just declared that they will ignore the result of a vote if it doesn't go the way they want.
Say at the next election they *do* get voted out. So what? What's to stop them declaring *again* that the voters didn't understand the issue, or were confused, and that they just declare that the *correct* result is that they've been voted back in.
If your votes are being thrown away, discarded or ignored, there *is* no point in voting. It's just a meaningless sham, dressed up to look like democracy to anyone who's only paying a little bit of attention. Sort of what a 3rd world dictatorship might do in the hope of getting on the good side of... well, I was going to say the USA, but perhaps the UN might be a better option given this very result.
This is the point where those 2nd amendment rights might start coming in useful...
"Authority and responsibility must be equal - else a balancing takes place as surely as current flows between points of unequal potential. To permit irresponsible authority is to sow disaster; to hold a man responsible for anything he does not control is to behave with blind idiocy."
Linux is well supported on at least 11 architectures. To claim that a piece of software "runs on Linux" with no further qualifications implies that it runs on them all.
If you're releasing a binary-blob-piece-of-shit that only runs on Linux/x86 (or, even worse, only Linux/x86-32), then you're not "supporting Linux" and should be fucking ashamed of lying so goddamn brazenly. You should at least have the balls to state "runs on Linux/x86-32" if that's what you mean.
(If you're relying on glibc extensions to the standard C library, you should really state GNU/Linux/x86-32, but that's another kettle of fish)
Caveats for running 'Sid': Right now, you'll probably be alright. When 'Etch' is released in few months time and becomes 'stable', 'Sid'/'unstable' will very likely become actually quite unstable for a while, and it might be worth moving to the new 'testing' distribution, whatever that is.
The reason for this is because 'Etch' is close to release. We're not in full freeze yet, but/some/ parts are frozen, and others are certainly crystalising a fair amount, so there's not *that* much going on in Etch. However, in order for updates to get to 'Etch', they have to spend some time in 'Sid' first. This means that nothing really radical is happening in 'Sid' either, so that small updates and bugfixes can be applied to current packages, spend some time in 'Sid' and then transition to 'Etch' without causing huge waves and potentially upping the Release Critical (RC) bug count.
In turn, this means that all the big breaking changes that are going to be held off between now and the end of the year will likely all hit 'Sid' at once after 'Etch' is released, and one or two are likely to cause some bad breakage. Certainly, shortly after 'Sarge' was released there were a whole bunch of awkward transitions (including 2 (two) C++ ABI transitions) that were quite painful for those running 'Sid'.
But, that's what 'Sid'/'unstable' is for. Catching the worst of the problems before they hit 'testing'.
So, while 'Sid' is "most likely fine" for now, be wary after 'Etch' is released. If you keep anything important on your system, it might be worth transitioning to 'testing' at that point.
Once upon a midnight dreary, while I pondered weak and weary, Over many a quaint and curious man page of forgotten lore, While I nodded, nearly napping, to page down I started tapping, On the case of laptop rapping, rapping to see one page more. `'Tis a kernel module,' said I, `understanding taps galore - Only this, and nothing more.'
That is the main cause of problems with windows. In order to develop on windows before.NET came along, you could pretty much guarantee you were doing COM programming of some kind.
In order to develop and test COM components, you need to be able to install them so they can be picked up by CoCreateInstance() and its brethren.
In order to install COM components, you need to write to the HKEY_LOCAL_MACHINE/[something I can't remember]/Classes registry key, which rightly requires Administrator priveliges.
Therefore, in order to develop COM components - i.e. in order to do any serious Windows development - you _needed_ to run as roo^H^H^HAdministrator.
The trouble is, if you're running as Administrator, you don't notice if you end up doing other things that also require Administrator privs. Like writing to other parts of HKLM, or the Program Files (or even System32) folder. None of that fails on your system, as you're Admin, so you never pick up on it.
Your code goes to testing. In order to install your COM components, the testers need to at least install as Administrator. If they forget to test as a limited user, they'll never notice that either.
You ship. Now everyone who uses your program needs to run as Admin. Think they'll change which account they're logged into just to run your program? Nope - they'll just give themselves Admin privs all the time.
In order to fix this, developers _need_ to be able to write, compile, install and run the programs they're developing as non-Administrators. Fortunately,.NET allows you to do this as you don't _have_ to install your stuff to the GAC, and you don't need to write to any privileged locations to be able to get stuff done. You should just be able to play in your hom^H^H^HDocuments and Settings folder and do everything from there.
Hmmm...with Linux, the only places that regular users can write to anyway is their home directory and/tmp. They need write access to those areas to be able to save stuff. Unfortunately, we probably can't stop them creating or downloading executables to those areas.
However, mount(8) has a great option - "noexec" - that can be used to prevent files from any partition being executed. If you put restricted users' home directories in/nxhome (no execute home) and mount/nxhome and/tmp as "noexec", that would probably do the trick.
DNS is a paid-for service; it's just not paid for by people doing the lookups. It's being paid for by the people who want DNS entries!
You don't _need_ a DNS entry to have a host on the internet, or even a website. Try going to http://66.35.250.150/. Buying a DNS entry is like buying a large ad in the business section of the telephone directory, or in the local free paper.
Sure, you could charge people for papers/phone directories/DNS queries/etc..., but the people buying the ads don't like it as it means less people will want to see the ad they've payed money to place, and the people using the paper/DNS don't tend to bother as they'll get the free one instead.
"The ultimate irony about interviews and statements made to the press is that Microsoft keeps claiming to be making a secure operating system, while Linus Torvalds keeps claiming to be striving for world domination."
Why is it that Slashdotters always say 'oh, this exploit just affects userland, no big deal'
Why is it that most people who trot out that line always assume that because a windows exploit can take down their OS, it isn't going to trash their home directory as well?
Also, it's a hell of a lot easier to restore a single user's files if the rest of the OS is still intact.
If your OS gets pwn3d, you can't trust it. At all. You know the r00tkit tech that Sony has recently been grilled about? It's called a r00tkit as if you have one it allows an attacker to keep r00t on your box without you knowing about it. So, if your OS dies, you need to wipe the lot and reinstall from scratch to be sure it's gone.
If you've been lucky enough to have installed your OS on a separate partition from your personal files, and none of your personal files have been touched (despite your OS getting hosed), then a reformat and reinstall of all your apps might only take you, oh, 2-3 hours?
If your OS is on the same partition as your personal stuff, you have to be careful about what you blow away, and things take longer.
If your personal files get trashed as well as your OS, well, you've got the 2-3 hours to restore the OS and all those apps, as before. Then you have all your personal files to restore. Then you have everyone elses personal files to restore. And they didn't even do anything bad! How pissed are they going to be if you've lost some of their work?
An exploit that just affects one user's personal files is a hell of a lot easier to recover than an exploit that affects everyone's personal files, and the OS you're accessing those files with.
Crikey. If you generally run as a user who has the privs to delete perl, apache, startup scripts, X, ftp, grep, ssh or anything of that ilk, please stop. You're just making it easier for the black-hats to 0wn j00!
Even if you don't care about the security of your boxen, I do. I don't want you spewing spam, DOS attacks or worms any further than they already are. While letting your user account get h4x0r3d might not prevent this completely (anyone can make an outbound connection to port 25, 80 or just do ICMP floods), giving the attackers r00t is definitely something none of us want.
However, because Setup downloads only those files that are necessary for your computer, this size can vary between 11 and 75 MB
What is a typical installation, and what is it based off of? I'm guessing that for IE6 SP1, "only those files that are neccessary" means that the 25MB is an average download for those working from a base of IE6.0. So, typically, upgrading from 6.0 to 6SP1 is a 25MB download.
What angle_slam was talking about in comment #9677091 was removing IE completely from the system as it's not used. So while you might be able to save 25MB by just deleting the parts of IE6 that are updated by SP1, I'm guessing that if you got rid of all of IE you could probably save around 100MB.
Up to 75MB for a browser! Like I mentioned - FireFox is less than 5MB from scratch starting with any base Win32 system you care to name. Any browser that could take up to 75MB to download is a fucking joke. Heck, the Mozilla Application Suite 1.7 is 12MB or so (again, starting with any Win32 system you care to name), and that includes a mail client and HTML editor. It's still half the size of the IE6 SP1 typical browser install, and a sixth of the monstrosity you might have to get if you're really unlucky.
10MB? Where do you get that? According to http://www.microsoft.com/windows/ie/downloads/crit ical/ie6sp1/default.mspx the IE6 SP1 download can be up to 75MB in size! Uncompressed, you could probably free up over 100MB if you could get rid of IE6 completely.
Firefox comes in at under 5MB compressed, and incorporates its own widget set in that.
If RMS were here to hear you say that, he'd kick your ass all the way from here to Albaquerque(sp?) and back again.
Cygwin is not Linux. Not in any way shape or form could it be considered Linux. The thing you think of as Linux - that RMS and the rest of the FSF would like you to call GNU/Linux - is a bunch of 3rd party tools running on top of hundreds of GNU utilities and libraries (including the GNU bash shell and the GNU C library (glibc), all compiled with the GNU Compiler Collection (GCC)) talking (via the aforementioned glibc) to the Linux kernel.
Now, you're taking Linux away from that, and making a C library (admittedly newlib instead of glibc) talk to Win32 instead, and recompiling all the GNU utilites (like bash) with the GNU compiler collection (GCC), and calling it "Cygwin Linux"?
Take a trip to the very bottom of the POSIX advocacy class.
Cygwin (or GNU/Cygwin if you insist on sharing the attribution with someone) is a UNIX(tm)-like environment. This is because it implements many parts of the shell, utilities and system calls specified in the Single Unix Specification (aka IEEE 1003.1, aka ISO 9945-{1,2,3}:2003, aka POSIX.1 and POSIX.2), as defined and downloadable from http://www.unix.org/. All systems that match the spec to a certain level of conformance (and possibly do other things like pay money and get certification?) can call themselves UNIX systems. Sytems that implement most of the spec or just haven't got around to certifying themselves yet call themselves UNIX-like systems. Linux is such a UNIX or UNIX-like system. As is *BSD. As is (or will be) GNU/Hurd. As is Solaris. As is HP-UX. As is AIX.
As is Cygwin. It implements a UNIX-like interface on the Win32 subsystem/kernel.
But Cygwin is not a Linux system any more than it is a SCO Unixware system, or a Solaris system, as it contains and uses zero code whatsoever from Linux. Linux is a kernel. It is only a kernel. It does little more than schedule processes, manage memory, provide mechanisms for talking to assorted bits of hardware, and provide a virtual filesystem for mounting tree-based structures into.[1] If you take that away, no matter what's left, it ain't Linux.
Heh. Not only did he do that, he returned the pointer into an int (which isn't guaranteed to work), and then failed to return any value at all if limit = 0 (which is always fun for the caller).
Burn your keyboard. And your C manuals. Take up Visual Basic or something:)
Won't work if javascript is not enabled or ads are not shown?
Hmmm... Not sure what it's supposed to look like, but I can see a table of the elements and browse to information about them fine using links, which has no javascript or image support.
No, but NVIDIA is not an ISV. It's a hardware company ferchrissakes!
It doesn't matter if I have a copy of their great software to render movies with if I don't have one of these whiz-bang cards to render it on. How on earth am I (or L33t J. Hax0r) going to copy a piece of hardware? The premise you quote is shaky at best (RedHat, among others, makes a profit from selling support & consulting services) but completely pointless when directed at a hardware company!
Look at IBM. They're making loads of money with Linux, not because of Linux sales or support or consulting, but because they're using it as a way to help sell their hardware.
Slashdot Mirror
Edgy and Feisty are only supported _by Ubuntu_ for 18 months. There's no reason Dell can't provide, or pay another 3rd party to provide, or even pay Ubuntu to provide, longer term support for the version of Ubuntu they ship. That's one of the beauties of Free Software - anyone can provide support for any bit of software, because anyone can go in there and fix bugs, even after upstream has stopped supporting that version.
Um....but these people have just declared that they will ignore the result of a vote if it doesn't go the way they want.
... well, I was going to say the USA, but perhaps the UN might be a better option given this very result.
Say at the next election they *do* get voted out. So what? What's to stop them declaring *again* that the voters didn't understand the issue, or were confused, and that they just declare that the *correct* result is that they've been voted back in.
If your votes are being thrown away, discarded or ignored, there *is* no point in voting. It's just a meaningless sham, dressed up to look like democracy to anyone who's only paying a little bit of attention. Sort of what a 3rd world dictatorship might do in the hope of getting on the good side of
This is the point where those 2nd amendment rights might start coming in useful...
That must mean that Duke Nukem Forever is going to ROCK!! ;)
(Less than 5 months until Apr 27 2007, the 10-year anniversary of the initial announcement)
http://kerneltrap.org/node/7184
"Authority and responsibility must be equal - else a balancing takes place as surely as current flows between points of unequal potential. To permit irresponsible authority is to sow disaster; to hold a man responsible for anything he does not control is to behave with blind idiocy."
-- Robert A. Heinlen, _Starship Troopers_
...this is Linux/x86-32 support.
Linux is well supported on at least 11 architectures. To claim that a piece of software "runs on Linux" with no further qualifications implies that it runs on them all.
If you're releasing a binary-blob-piece-of-shit that only runs on Linux/x86 (or, even worse, only Linux/x86-32), then you're not "supporting Linux" and should be fucking ashamed of lying so goddamn brazenly. You should at least have the balls to state "runs on Linux/x86-32" if that's what you mean.
(If you're relying on glibc extensions to the standard C library, you should really state GNU/Linux/x86-32, but that's another kettle of fish)
Caveats for running 'Sid': Right now, you'll probably be alright. When 'Etch' is released in few months time and becomes 'stable', 'Sid'/'unstable' will very likely become actually quite unstable for a while, and it might be worth moving to the new 'testing' distribution, whatever that is.
/some/ parts are frozen, and others are certainly crystalising a fair amount, so there's not *that* much going on in Etch. However, in order for updates to get to 'Etch', they have to spend some time in 'Sid' first. This means that nothing really radical is happening in 'Sid' either, so that small updates and bugfixes can be applied to current packages, spend some time in 'Sid' and then transition to 'Etch' without causing huge waves and potentially upping the Release Critical (RC) bug count.
The reason for this is because 'Etch' is close to release. We're not in full freeze yet, but
In turn, this means that all the big breaking changes that are going to be held off between now and the end of the year will likely all hit 'Sid' at once after 'Etch' is released, and one or two are likely to cause some bad breakage. Certainly, shortly after 'Sarge' was released there were a whole bunch of awkward transitions (including 2 (two) C++ ABI transitions) that were quite painful for those running 'Sid'.
But, that's what 'Sid'/'unstable' is for. Catching the worst of the problems before they hit 'testing'.
So, while 'Sid' is "most likely fine" for now, be wary after 'Etch' is released. If you keep anything important on your system, it might be worth transitioning to 'testing' at that point.
Also read this document.
Once upon a midnight dreary, while I pondered weak and weary,
Over many a quaint and curious man page of forgotten lore,
While I nodded, nearly napping, to page down I started tapping,
On the case of laptop rapping, rapping to see one page more.
`'Tis a kernel module,' said I, `understanding taps galore -
Only this, and nothing more.'
"What about developers?"
.NET came along, you could pretty much guarantee you were doing COM programming of some kind.
.NET allows you to do this as you don't _have_ to install your stuff to the GAC, and you don't need to write to any privileged locations to be able to get stuff done. You should just be able to play in your hom^H^H^HDocuments and Settings folder and do everything from there.
That is the main cause of problems with windows. In order to develop on windows before
In order to develop and test COM components, you need to be able to install them so they can be picked up by CoCreateInstance() and its brethren.
In order to install COM components, you need to write to the HKEY_LOCAL_MACHINE/[something I can't remember]/Classes registry key, which rightly requires Administrator priveliges.
Therefore, in order to develop COM components - i.e. in order to do any serious Windows development - you _needed_ to run as roo^H^H^HAdministrator.
The trouble is, if you're running as Administrator, you don't notice if you end up doing other things that also require Administrator privs. Like writing to other parts of HKLM, or the Program Files (or even System32) folder. None of that fails on your system, as you're Admin, so you never pick up on it.
Your code goes to testing. In order to install your COM components, the testers need to at least install as Administrator. If they forget to test as a limited user, they'll never notice that either.
You ship. Now everyone who uses your program needs to run as Admin. Think they'll change which account they're logged into just to run your program? Nope - they'll just give themselves Admin privs all the time.
In order to fix this, developers _need_ to be able to write, compile, install and run the programs they're developing as non-Administrators. Fortunately,
Hmmm...with Linux, the only places that regular users can write to anyway is their home directory and /tmp. They need write access to those areas to be able to save stuff. Unfortunately, we probably can't stop them creating or downloading executables to those areas.
/nxhome (no execute home) and mount /nxhome and /tmp as "noexec", that would probably do the trick.
However, mount(8) has a great option - "noexec" - that can be used to prevent files from any partition being executed. If you put restricted users' home directories in
"DNS is free right now."
No it isn't! *slap*
DNS is a paid-for service; it's just not paid for by people doing the lookups. It's being paid for by the people who want DNS entries!
You don't _need_ a DNS entry to have a host on the internet, or even a website. Try going to http://66.35.250.150/. Buying a DNS entry is like buying a large ad in the business section of the telephone directory, or in the local free paper.
Sure, you could charge people for papers/phone directories/DNS queries/etc..., but the people buying the ads don't like it as it means less people will want to see the ad they've payed money to place, and the people using the paper/DNS don't tend to bother as they'll get the free one instead.
:)
Do people still get them? I thought everyone had adblock installed.
"The ultimate irony about interviews and statements made to the press is that Microsoft keeps claiming to be making a secure operating system, while Linus Torvalds keeps claiming to be striving for world domination."
-- Unknown.
Why is it that Slashdotters always say 'oh, this exploit just affects userland, no big deal'
Why is it that most people who trot out that line always assume that because a windows exploit can take down their OS, it isn't going to trash their home directory as well?
Also, it's a hell of a lot easier to restore a single user's files if the rest of the OS is still intact.
If your OS gets pwn3d, you can't trust it. At all. You know the r00tkit tech that Sony has recently been grilled about? It's called a r00tkit as if you have one it allows an attacker to keep r00t on your box without you knowing about it. So, if your OS dies, you need to wipe the lot and reinstall from scratch to be sure it's gone.
If you've been lucky enough to have installed your OS on a separate partition from your personal files, and none of your personal files have been touched (despite your OS getting hosed), then a reformat and reinstall of all your apps might only take you, oh, 2-3 hours?
If your OS is on the same partition as your personal stuff, you have to be careful about what you blow away, and things take longer.
If your personal files get trashed as well as your OS, well, you've got the 2-3 hours to restore the OS and all those apps, as before. Then you have all your personal files to restore. Then you have everyone elses personal files to restore. And they didn't even do anything bad! How pissed are they going to be if you've lost some of their work?
An exploit that just affects one user's personal files is a hell of a lot easier to recover than an exploit that affects everyone's personal files, and the OS you're accessing those files with.
That's why slashdotters say that.
I'm betting they communicate with a belt-mounted (or in pocket) Perilometer, and turn completely black at the first sign of danger.
The all-in-one Joo Janta 200 Super-Chromatic version won't be out 'til next year though.
He must be new here.
STOP RUNNING AS ADMIN/ROOT!
Crikey. If you generally run as a user who has the privs to delete perl, apache, startup scripts, X, ftp, grep, ssh or anything of that ilk, please stop. You're just making it easier for the black-hats to 0wn j00!
Even if you don't care about the security of your boxen, I do. I don't want you spewing spam, DOS attacks or worms any further than they already are. While letting your user account get h4x0r3d might not prevent this completely (anyone can make an outbound connection to port 25, 80 or just do ICMP floods), giving the attackers r00t is definitely something none of us want.
However, because Setup downloads only those files that are necessary for your computer, this size can vary between 11 and 75 MB
What is a typical installation, and what is it based off of? I'm guessing that for IE6 SP1, "only those files that are neccessary" means that the 25MB is an average download for those working from a base of IE6.0. So, typically, upgrading from 6.0 to 6SP1 is a 25MB download.
What angle_slam was talking about in comment #9677091 was removing IE completely from the system as it's not used. So while you might be able to save 25MB by just deleting the parts of IE6 that are updated by SP1, I'm guessing that if you got rid of all of IE you could probably save around 100MB.
Up to 75MB for a browser! Like I mentioned - FireFox is less than 5MB from scratch starting with any base Win32 system you care to name. Any browser that could take up to 75MB to download is a fucking joke. Heck, the Mozilla Application Suite 1.7 is 12MB or so (again, starting with any Win32 system you care to name), and that includes a mail client and HTML editor. It's still half the size of the IE6 SP1 typical browser install, and a sixth of the monstrosity you might have to get if you're really unlucky.
10MB? Where do you get that? According to http://www.microsoft.com/windows/ie/downloads/crit ical/ie6sp1/default.mspx the IE6 SP1 download can be up to 75MB in size! Uncompressed, you could probably free up over 100MB if you could get rid of IE6 completely.
Firefox comes in at under 5MB compressed, and incorporates its own widget set in that.
Cygwin Linux?!?
If RMS were here to hear you say that, he'd kick your ass all the way from here to Albaquerque(sp?) and back again.
Cygwin is not Linux. Not in any way shape or form could it be considered Linux. The thing you think of as Linux - that RMS and the rest of the FSF would like you to call GNU/Linux - is a bunch of 3rd party tools running on top of hundreds of GNU utilities and libraries (including the GNU bash shell and the GNU C library (glibc), all compiled with the GNU Compiler Collection (GCC)) talking (via the aforementioned glibc) to the Linux kernel.
Now, you're taking Linux away from that, and making a C library (admittedly newlib instead of glibc) talk to Win32 instead, and recompiling all the GNU utilites (like bash) with the GNU compiler collection (GCC), and calling it "Cygwin Linux"?
Take a trip to the very bottom of the POSIX advocacy class.
Cygwin (or GNU/Cygwin if you insist on sharing the attribution with someone) is a UNIX(tm)-like environment. This is because it implements many parts of the shell, utilities and system calls specified in the Single Unix Specification (aka IEEE 1003.1, aka ISO 9945-{1,2,3}:2003, aka POSIX.1 and POSIX.2), as defined and downloadable from http://www.unix.org/. All systems that match the spec to a certain level of conformance (and possibly do other things like pay money and get certification?) can call themselves UNIX systems. Sytems that implement most of the spec or just haven't got around to certifying themselves yet call themselves UNIX-like systems. Linux is such a UNIX or UNIX-like system. As is *BSD. As is (or will be) GNU/Hurd. As is Solaris. As is HP-UX. As is AIX.
As is Cygwin. It implements a UNIX-like interface on the Win32 subsystem/kernel.
But Cygwin is not a Linux system any more than it is a SCO Unixware system, or a Solaris system, as it contains and uses zero code whatsoever from Linux. Linux is a kernel. It is only a kernel. It does little more than schedule processes, manage memory, provide mechanisms for talking to assorted bits of hardware, and provide a virtual filesystem for mounting tree-based structures into.[1] If you take that away, no matter what's left, it ain't Linux.
K.
[1] Intentional understatment.
Heh. Not only did he do that, he returned the pointer into an int (which isn't guaranteed to work), and then failed to return any value at all if limit = 0 (which is always fun for the caller).
:)
Burn your keyboard. And your C manuals. Take up Visual Basic or something
Won't work if javascript is not enabled or ads are not shown?
Hmmm... Not sure what it's supposed to look like, but I can see a table of the elements and browse to information about them fine using links, which has no javascript or image support.
No, but NVIDIA is not an ISV. It's a hardware company ferchrissakes!
It doesn't matter if I have a copy of their great software to render movies with if I don't have one of these whiz-bang cards to render it on. How on earth am I (or L33t J. Hax0r) going to copy a piece of hardware? The premise you quote is shaky at best (RedHat, among others, makes a profit from selling support & consulting services) but completely pointless when directed at a hardware company!
Look at IBM. They're making loads of money with Linux, not because of Linux sales or support or consulting, but because they're using it as a way to help sell their hardware.