I don't want that 911 tracing. Just what we need, another easy way for the government to keep tabs on us
Retard, you have to call 911 with your phone for them to trace your location. The only reason to call 911 is to call for emergency services. If you need help for yourself, they need to know where you are. If you're calling for someone else, odds are pretty good that you're geographically close to that other person.
It's just like those automated toll collectors, they are now using that to give speeding tickets (by timing your time between pay points) and for tracking movement in legal cases.
Any articles or public announcements about the speeding tickets? Why are you so upset that people breaking the law are more likely to get caught? If someone is going 120 down the highway, wouldn't you want to know they're going to be punished for driving wrecklessly? I know I would, and you have to enforce the law uniformly.
The data is already there for them to track, they just needs the proper warrants and such to be able to use the data. (And I don't really want to debate the due process and search/seizure laws right now, that's offtopic)
Unless companies allow you to go month by month, you can't switch b/c their service is bad.
Sprint odesn't advertise it, but you don't need a contract with them. They just charge you $10 a month extra without it. I know this first hand after calling to complain about the $10 charge when I had, in fact, signed an agreement. Also, after your agreement term has passed, they just keep billing you at the same rate, no sudden surge in sales calls to sign up for a new plan or anything. I'm pretty happy with it. I just wish they has a selection of phones without antennas.
Seriously, how many people had the opportunity to switch, but would not b/c they would lose their number.
Me, for one, when I had to replace my phone and Sprint wasn't offering any deals for phone upgrades to existing customers. If I didn't have to buy new phones, I would consider switching to Verizon or T-Mobile, now that I can keep my number.
noone of the spams I ever received contained such.
Not all of us are as fortunate as you then.
So, which nasty people did you give your email address to?
studies have shown that having you email address in unaltered text on a web site is the most sure-fire way to get on spam lists. So any web archive of USENET or mailing lists which does not strip the address out of the headers is like a candy store. Even if they do strip the headers, lots of people put their address in an email signature, which will get displayed in the archive.
Remember that Sun is a hardware manufacturer. Meaning, if you're rolling out JDS, they're probably hoping you're rolling it out on their hardware.
Granted, but Sun is basing JDS on Linux, which should already support the card.
Also note that there is a fix for it
The review said the only solution was to build a custom kernel. I would have to assume that SuSE, and in turn Sun, distributes a customized kernel config, which could mean you must lose those customizations to fix the NIC. Unless Eugenia was wrong and there are updated packages from SuSE/Sun, which wouldn't surprise me too much.
How minor would you consider it if your company had 500 desktops with that specific NIC? Would you shell out the cash for new NICs? Would you be happy rolling a custom built kernel? The Realtek cards are pretty well supoprted by Linux, and SuSE has had this bug for almost a year (according to the review). If Microsoft let a buggy driver sit like that people would screm bloody murder.
I agree that she's not very good at accurately portraying the severity or impact of bugs, I don't think this is such a case.
You're right about the emulator vs implementation argument, but it's still pretty slow. The problem with Wine (plain Wine, not WineX or Crossover, they might have better implementations of common DLLs, never used either so I can't comment) is that you still have to own a copy of windows to legeally have copies of the DLLs that most interesting applications need. At that point, you're better off dual-booting (if you only have one machine) and avoiding the performance hit. Anecdotally, I notice the performance hit running Kazaa Lite K++ in Wine versus in Windows.
There were two ways to interpret the original question: how do I mount the network shared NTFS drives from my Windows box on my Linux box; or how do I mount the Windows partitions on my Linux box so I can share them over the network with Samba. The indication was there that the physical discs with the NTFS partitions would be in the Linux box, so the kernel would have to understand NTFS to mount them. Then Samba could be used to share them over the network.
That should be sending me a flag that this is just a troll or flamebait, but I'm biting anyway. I don't have the money to buy an extra machine so I can run Linux and my wife can run Windows. And I'm not such a zealot as to make her use Linux for tasks that she finds easier in Windows. There is no Photoshop for Linux, and the only legal ways to run Photoshop in Linux end up meaning I have to have a copy of Windows. (VMWare + Windows, Bochs + Windows, Wine + Windows DLLs). So if I'm already paying for Windows, then I may as well dualboot it and avoid the performance hit of VMWare/Bochs/Wine.
Watergate - If that is the worst thing you can think of the U.S. government subjecting its citizens to, I think that says something about our society. Government officials were responsible for espionage, they were caught, and most of them punished. I think the DMCA, the PARIOT Act, and sending troops to 'war' without Congress declaring war are far worse. Those fly in the face of what our Consitution grants the government power over.
Iran/Guantanamo - I'm not denying that there was any wrongdoing, but how does that compare with the systematic slaughter of citizens to maintain political control?
Seriously, I'm not saying the U.S. government is perfect, but I hardly see how anything you list compares with what happens under millitary dictatorships and the like. British occupation of India and Apartheid in South Africa are just two things I can think of done by a 'benevolent' government which outhsine your examples. Maybe that's just the propaganda talking though, but I spent most of my childhood in England so I doubt it.
It's an appeal, it's not a second trial, it's not Double Jeopardy. Read the other responses to your posts. It's just a fresh set of eyes looking at the original proceedings to see if any mistakes were made. The U.S. allows appeals too, get over it. A free society recognises that both sides must have equal rights under the law. If the defendant can appeal then so can the plaintiff. Or perhaps you'd prefer a system where one side has more rights than the other, champion of freedom that you appear to be.
Yeah but the question is can you pass application specific build parameters
I can't say for certain, but I would have to assume so. If dpkg is anything like rpm it has a simple config file to set such parameters. I know you can pass options on the command line when building rpms though.
Threading - Why would I watch a DVD on a dev machine or server? You said Linux was a bad server because it can't do multiple compiles. I said I have no trouble doing concurrent builds, so you tell me to watch a DVD and compile software. Now you're mixing desktop and server usage to say FreeBSD is a better server. I don't deny the latency issues you bring up, but I do take issue with the examples you use to illustrate your point.
Hardware support - You said, and I quote, "BSD isnt meant for desktop, its meant for doing work, I dont need my nVidia card to be supported to do email and crap." I gave an example of where some people need 3d hardware support for "doing work." You respond with FreeBSD is meant to be a work horse, not a graphics workstation. I'm sure you didn't mean to minimize egineers in the oil industry, but you kinda did. Is compiling software magically more computationally difficult than running geological simulations and plotting fault lines and stuff like that? People doing real work need 3d hardware support and FreeBSD doesn't offer it. The developers have different goals, and the users have different needs.
Security - different design goals, different needs, yadda yadda.
License - You totally missed my point. The GPL allows me different freedoms as a publisher than the BSD license. The BSD license offers me more freedoms as a consumer than the GPL. But who cares? That's why we have more than one OSS license to choose from.
In case I haven't made it clear yet, my impression of your statements is that you don't accept the fact that some people need things FreeBSD doesn't offer, just like some people need things that Linux doesn't offer. I'm not saying that Linux is better, I'm just trying to say that FreeBSD isn't better either.
the current 2.4.18 kernel image for Debian stable has been kept current with security patches. It is already patched for this exploit. The same can be said for Red Hat kernel images. The kernel you get by building Debian or Red Hat sources is NOT the kernel you get from kernel.org. When Red Hat and Debian say stable, they mean the kernel API doesn't change. So they take security patches and rewrite them to work with the older kernels without changing the APIs and ABIs available to programmers.
Most Linux admins won't deploy a new kernel right away either. They will, however get the source diffs and patch the kernel sources that the running kernels were built off of, and run the patched kernel. Lots of Linux admins also have sandbox machines to try new patches before deploying to production servers. I know I sound like a fanboy in your ears with this next statement, but there is no way to be sure what is affected by a MS patch other than to install it and test it. With a Linux kernel patch, you look at the code, apply it, and deploy it. Anyone just running binary updates from their vendor should have the same trust in MS patches that they do in patches from their Linux vendor. (Note that I'm not implying doing see is either good or bad)
Perhaps I should have included a disclaimer that I don't agree with it. I just thought the quote was more relevant to the thread than the quote presented in the post I was replying to.;)
Mandrake started as RH plus KDE. For the longest time, MDK and RH were binary compatible on pentium plus machines
I wonder if there were any other distros based off the same RH releases. One could then show that packages that common base release would work just fine on both the derivative distributions (just as base debian packages work on debian derived distros).:)
Considering that MDK was the first LSB compliant distro, so if anyone is at fault it would be RH.
In priniciple I agree, but does LSB dictate the generic packages names, and thus how to name dependencies? If not, then it's not Red Hat's fault that Mandrake changed the package names and built packages dependent on those changed names. RPM supports file dependencies as well as generic package name dependencies. So vendors should either standardize package names, or use LSB compliant file paths for their dependencies. In absence of those two, all vendors are equally at fault for creating vendor-specific binary packages.
You arent fetching sources, you're fetching compiled binaries. The beauty of FreeBSD is you can optimize your system
The APT system supports source repositories. The official Debian archives offer the source to every package, so it is entirely possible to install the base system with the installer, and then use nothing but locally built packages using apt/dpkg.
Linux can not handle multiple compiles at once, hence its not really meant as a server OS
Unless you intentionally left 'build' out of that sentence before 'server' this is a stupid statement. Since when is it a good idea to have a compiler on your server? Even if you did specifically mean build server, in most cases that just means running remote shells to do the compiling (like rsh or ssh) and not some spiffy server app that receives build requests from a build client (yes I know there are distributed build systems like Rational's clearmake, but that's not the norm). Besides all that it can handle multiple compiles. I have a build host which runs at least 9 builds at the same time every day, works just fine. The applications being built are used by major oil companies for geophysical mapping and surveying, so they're not exactly trivial, either.
As for the drivers issue, big deal, BSD isnt meant for desktop, its meant for doing work, I dont need my nVidia card to be supported to do email and crap.
Some people do need the latest nVidia cards to be supported, so if FreeBSD doesn't support it, it's not a viable solution. People doing graphical modeling software in the oil industry need good 3d hardware support. I work for such a company (subsidiary of Haliburton), and we are getting pressure from Shell to support Red Hat Enterprise Linux 3.0 (we already wupport 2.1, but they want to upgrade).
Different licenses fill different needs. With the BSD license I am not free to ensure that people benefiting from my work return anything to the community.
To say that either system is always faster or more stable or more secure is pure nonsense. anyone can misconfigure a system and make it perform like shit. Anyone can tweak a system and make it scream. Developers of different projects have different goals, so they get different results.
Once you learn that different people have different needs, and different tools must be used to maximize efficiency, you might realise that it's pointless to bicker about which OSS platfrom is superior. They are all simultaneously superior and inferior to one another.
The beauty of Debian installs is that.deb's are pretty much universal. I can install Knoppix or Mepis and link straight to the Debian package sources, and everything pretty much "just works" when I do apt-get update / upgrade. Not so with suse / rh, which is one of the big reasons I won't go back to RPM.
That's a terrible comparison. You're comparing disitrubtions (knoppix and mepis) based on a common disitribution (debian) to distributions (Suse and Red Hat) based on a common packaging format (RPM). Find a distribution which uses dpkg but is not based on Debian. Then see how universal the Debian.deb's are on it, or vice versa. Until Fedora Core, there has been no framework for entire disitributions built off of a Red Hat offering. Now we see the efforts of 3rd party pacakge maintainers like fedora.us, livna.org, freshrpms.net, AtRPMS, and Dag Wiers coming together in a way that will hopefully lend itself to having Fedora Core as a reference disitribution, and other organisations building customized releases that use the base repositories and a few 3rd party repos and some custom themes and stuff.
It's the reason people say things like, "Slashdot isn't pro-Linux, it's anti-M$."
I find the quote "FreeBSD is for people who love Unix, Linux is for people who hate Microsoft," a little more apropos. It not only emphasizes the OSS vs Proprietary sentiment that sparked this thread, it illustrates the OSS infighting commented on in the article too.
Slashdot Mirror
Retard, you have to call 911 with your phone for them to trace your location. The only reason to call 911 is to call for emergency services. If you need help for yourself, they need to know where you are. If you're calling for someone else, odds are pretty good that you're geographically close to that other person.
It's just like those automated toll collectors, they are now using that to give speeding tickets (by timing your time between pay points) and for tracking movement in legal cases.
Any articles or public announcements about the speeding tickets? Why are you so upset that people breaking the law are more likely to get caught? If someone is going 120 down the highway, wouldn't you want to know they're going to be punished for driving wrecklessly? I know I would, and you have to enforce the law uniformly.
The data is already there for them to track, they just needs the proper warrants and such to be able to use the data. (And I don't really want to debate the due process and search/seizure laws right now, that's offtopic)
so don't ask to have your old number moved. It's not an automatic service, you have to request it, otherwise you just get a fresh new number.
Sprint odesn't advertise it, but you don't need a contract with them. They just charge you $10 a month extra without it. I know this first hand after calling to complain about the $10 charge when I had, in fact, signed an agreement. Also, after your agreement term has passed, they just keep billing you at the same rate, no sudden surge in sales calls to sign up for a new plan or anything. I'm pretty happy with it. I just wish they has a selection of phones without antennas.
Seriously, how many people had the opportunity to switch, but would not b/c they would lose their number.
Me, for one, when I had to replace my phone and Sprint wasn't offering any deals for phone upgrades to existing customers. If I didn't have to buy new phones, I would consider switching to Verizon or T-Mobile, now that I can keep my number.
Not all of us are as fortunate as you then.
So, which nasty people did you give your email address to?
studies have shown that having you email address in unaltered text on a web site is the most sure-fire way to get on spam lists. So any web archive of USENET or mailing lists which does not strip the address out of the headers is like a candy store. Even if they do strip the headers, lots of people put their address in an email signature, which will get displayed in the archive.
Granted, but Sun is basing JDS on Linux, which should already support the card.
Also note that there is a fix for it
The review said the only solution was to build a custom kernel. I would have to assume that SuSE, and in turn Sun, distributes a customized kernel config, which could mean you must lose those customizations to fix the NIC. Unless Eugenia was wrong and there are updated packages from SuSE/Sun, which wouldn't surprise me too much.
How minor would you consider it if your company had 500 desktops with that specific NIC? Would you shell out the cash for new NICs? Would you be happy rolling a custom built kernel? The Realtek cards are pretty well supoprted by Linux, and SuSE has had this bug for almost a year (according to the review). If Microsoft let a buggy driver sit like that people would screm bloody murder.
I agree that she's not very good at accurately portraying the severity or impact of bugs, I don't think this is such a case.
I've never once had any data loss using partition magic, in 5 years of use.
make the one-time investment in partition magic. boot off the CD and convert it from ntfs to ext2 or ext3.
You're right about the emulator vs implementation argument, but it's still pretty slow. The problem with Wine (plain Wine, not WineX or Crossover, they might have better implementations of common DLLs, never used either so I can't comment) is that you still have to own a copy of windows to legeally have copies of the DLLs that most interesting applications need. At that point, you're better off dual-booting (if you only have one machine) and avoiding the performance hit. Anecdotally, I notice the performance hit running Kazaa Lite K++ in Wine versus in Windows.
There were two ways to interpret the original question: how do I mount the network shared NTFS drives from my Windows box on my Linux box; or how do I mount the Windows partitions on my Linux box so I can share them over the network with Samba. The indication was there that the physical discs with the NTFS partitions would be in the Linux box, so the kernel would have to understand NTFS to mount them. Then Samba could be used to share them over the network.
That should be sending me a flag that this is just a troll or flamebait, but I'm biting anyway. I don't have the money to buy an extra machine so I can run Linux and my wife can run Windows. And I'm not such a zealot as to make her use Linux for tasks that she finds easier in Windows. There is no Photoshop for Linux, and the only legal ways to run Photoshop in Linux end up meaning I have to have a copy of Windows. (VMWare + Windows, Bochs + Windows, Wine + Windows DLLs). So if I'm already paying for Windows, then I may as well dualboot it and avoid the performance hit of VMWare/Bochs/Wine.
Iran/Guantanamo - I'm not denying that there was any wrongdoing, but how does that compare with the systematic slaughter of citizens to maintain political control?
Seriously, I'm not saying the U.S. government is perfect, but I hardly see how anything you list compares with what happens under millitary dictatorships and the like. British occupation of India and Apartheid in South Africa are just two things I can think of done by a 'benevolent' government which outhsine your examples. Maybe that's just the propaganda talking though, but I spent most of my childhood in England so I doubt it.
It's an appeal, it's not a second trial, it's not Double Jeopardy. Read the other responses to your posts. It's just a fresh set of eyes looking at the original proceedings to see if any mistakes were made. The U.S. allows appeals too, get over it. A free society recognises that both sides must have equal rights under the law. If the defendant can appeal then so can the plaintiff. Or perhaps you'd prefer a system where one side has more rights than the other, champion of freedom that you appear to be.
I can't say for certain, but I would have to assume so. If dpkg is anything like rpm it has a simple config file to set such parameters. I know you can pass options on the command line when building rpms though.
Hardware support - You said, and I quote, "BSD isnt meant for desktop, its meant for doing work, I dont need my nVidia card to be supported to do email and crap." I gave an example of where some people need 3d hardware support for "doing work." You respond with FreeBSD is meant to be a work horse, not a graphics workstation. I'm sure you didn't mean to minimize egineers in the oil industry, but you kinda did. Is compiling software magically more computationally difficult than running geological simulations and plotting fault lines and stuff like that? People doing real work need 3d hardware support and FreeBSD doesn't offer it. The developers have different goals, and the users have different needs.
Security - different design goals, different needs, yadda yadda.
License - You totally missed my point. The GPL allows me different freedoms as a publisher than the BSD license. The BSD license offers me more freedoms as a consumer than the GPL. But who cares? That's why we have more than one OSS license to choose from.
In case I haven't made it clear yet, my impression of your statements is that you don't accept the fact that some people need things FreeBSD doesn't offer, just like some people need things that Linux doesn't offer. I'm not saying that Linux is better, I'm just trying to say that FreeBSD isn't better either.
the current 2.4.18 kernel image for Debian stable has been kept current with security patches. It is already patched for this exploit. The same can be said for Red Hat kernel images. The kernel you get by building Debian or Red Hat sources is NOT the kernel you get from kernel.org. When Red Hat and Debian say stable, they mean the kernel API doesn't change. So they take security patches and rewrite them to work with the older kernels without changing the APIs and ABIs available to programmers.
Most Linux admins won't deploy a new kernel right away either. They will, however get the source diffs and patch the kernel sources that the running kernels were built off of, and run the patched kernel. Lots of Linux admins also have sandbox machines to try new patches before deploying to production servers. I know I sound like a fanboy in your ears with this next statement, but there is no way to be sure what is affected by a MS patch other than to install it and test it. With a Linux kernel patch, you look at the code, apply it, and deploy it. Anyone just running binary updates from their vendor should have the same trust in MS patches that they do in patches from their Linux vendor. (Note that I'm not implying doing see is either good or bad)
Perhaps I should have included a disclaimer that I don't agree with it. I just thought the quote was more relevant to the thread than the quote presented in the post I was replying to. ;)
I wonder if there were any other distros based off the same RH releases. One could then show that packages that common base release would work just fine on both the derivative distributions (just as base debian packages work on debian derived distros). :)
Considering that MDK was the first LSB compliant distro, so if anyone is at fault it would be RH.
In priniciple I agree, but does LSB dictate the generic packages names, and thus how to name dependencies? If not, then it's not Red Hat's fault that Mandrake changed the package names and built packages dependent on those changed names. RPM supports file dependencies as well as generic package name dependencies. So vendors should either standardize package names, or use LSB compliant file paths for their dependencies. In absence of those two, all vendors are equally at fault for creating vendor-specific binary packages.
The APT system supports source repositories. The official Debian archives offer the source to every package, so it is entirely possible to install the base system with the installer, and then use nothing but locally built packages using apt/dpkg.
Unless you intentionally left 'build' out of that sentence before 'server' this is a stupid statement. Since when is it a good idea to have a compiler on your server? Even if you did specifically mean build server, in most cases that just means running remote shells to do the compiling (like rsh or ssh) and not some spiffy server app that receives build requests from a build client (yes I know there are distributed build systems like Rational's clearmake, but that's not the norm). Besides all that it can handle multiple compiles. I have a build host which runs at least 9 builds at the same time every day, works just fine. The applications being built are used by major oil companies for geophysical mapping and surveying, so they're not exactly trivial, either.
As for the drivers issue, big deal, BSD isnt meant for desktop, its meant for doing work, I dont need my nVidia card to be supported to do email and crap.
Some people do need the latest nVidia cards to be supported, so if FreeBSD doesn't support it, it's not a viable solution. People doing graphical modeling software in the oil industry need good 3d hardware support. I work for such a company (subsidiary of Haliburton), and we are getting pressure from Shell to support Red Hat Enterprise Linux 3.0 (we already wupport 2.1, but they want to upgrade).
Different licenses fill different needs. With the BSD license I am not free to ensure that people benefiting from my work return anything to the community.
To say that either system is always faster or more stable or more secure is pure nonsense. anyone can misconfigure a system and make it perform like shit. Anyone can tweak a system and make it scream. Developers of different projects have different goals, so they get different results.
Once you learn that different people have different needs, and different tools must be used to maximize efficiency, you might realise that it's pointless to bicker about which OSS platfrom is superior. They are all simultaneously superior and inferior to one another.
That's a terrible comparison. You're comparing disitrubtions (knoppix and mepis) based on a common disitribution (debian) to distributions (Suse and Red Hat) based on a common packaging format (RPM). Find a distribution which uses dpkg but is not based on Debian. Then see how universal the Debian .deb's are on it, or vice versa. Until Fedora Core, there has been no framework for entire disitributions built off of a Red Hat offering. Now we see the efforts of 3rd party pacakge maintainers like fedora.us, livna.org, freshrpms.net, AtRPMS, and Dag Wiers coming together in a way that will hopefully lend itself to having Fedora Core as a reference disitribution, and other organisations building customized releases that use the base repositories and a few 3rd party repos and some custom themes and stuff.
I find the quote "FreeBSD is for people who love Unix, Linux is for people who hate Microsoft," a little more apropos. It not only emphasizes the OSS vs Proprietary sentiment that sparked this thread, it illustrates the OSS infighting commented on in the article too.
dust .... wind ....dude
both. Embedded Linux is an attractive host OS, and if that's the target platform, why not develop on a Linux box too?