You mentioned installing Linux. How many average users do you know that installed their version of windows rather than bought it with it on there already? Most of those people wouldn't be able to install windows. The fact is, for so called "n00b" distro's, the installion process is in fact easier and quicker than windows. Some such as Lindows and Mandrake can be installed onto a new computer with just a couple of steps and nothing more than pressing next. PC's bundled with Linux and linux friendly hardware alieve this problem completely.
RPM's are shitty and are the cause of more problems than they are worth. RPM's should have been done away with long ago. I remember as a noob, same problems. Could never install RPM's because of dependencys. I think distro's should adopt a portage style package system. Compiling from source takes longer, but it takes care of idiotic linking issues that RPM's create.
I think most of Linux's entry level issues will be solved when OEM's start shipping it more included. The reason Linux on the desktop at the office is a great canidate is because it's installed for the workers and they've got someone to help them get past their entry level issues. Which are minor, but can be a big deal to new users.
I may be the only one to feel this way... but haven't the completely unrelated SCO jokes gotten a little old? They are completely offtopic 99% of the time and most of them are just "Where do I send my 699 dollars".
I am willing to bet they've done _something_. For example, many times I tried to connect and my first try was immediatly stopped by differing error messages "This document contains no data" or "You don't have access to this resource". Then on a second attempt I am connect. Pretty much 9 times out of 10. Also, the initial connection to the main page might be slow; But once I'm on the page it seems fine. My belief is maybe they are purposly putting a latency on the first connection and then using keep-alive and allowing those conections more bandwidth. This would be similiar to a tar pit in that it's supposed to keep the client in a state of waiting as long as possible. The longer it waits, the less DoSing it can do. Just my two cents.
I think it's really dumb so many companies are backing gnome when, it's really an inferior window manager. First of all, it's a package maintainers nightmare. Gnome is split up into eleventy twelve bazillion packages. While Kde really only needs QT, Kdelibs, and Kdebase to get up and running.
QT is more mature, more stable, and simple a better toolkit than Gtk. Gtk is much more of a pain in the ass to program in than QT.
Gnome was a project that came out of RedHat... 'nuff said;-)
KDE has more to it. Things like KHtml and the kde regular expression library. A browser that's much more integrated into the wm than Galeon (have you tried the latest KDE rc? Konq opens faster than IE on windows). KWallet is awesome.
Not saying Gnome should go away. It's got some good individual packages. IIRC, Gnome is responsible for the XML related packages for linux. But why are so many supporting Gnome while ignoring KDE when it is such a great wm.
The problem isn't autoconf. Autoconf is a hack to compensate for the defencies of Automake. And the basic idea of needing a scriping/macro language to figure out what packages are installed and how the system is laid out is going to have to be there as long there is no standard file(s) that contains all package and system information. I think another language besides m4 should be used (perl or python?). In fact, tools looking to replace autoconf do in fact use python (haven't seen any perl yet though). I think a lot of people fail to realize that most OSS comes in the form of./configure && make && make install
Even those.deb,.rpm,.tgz binary packages are usually made with binaries that were compiled with the help of autoconf. Even gentoo's ebuild system uses autoconf.
The argument I hear the most, without a doubt "Windows gets more viruii because it's more popular". I call bullshit! I know it's bullshit because of Apache. Apache, by almost any web server survey, has at least as many servers as IIS (netcraft says between 2x and 3x, but let's say just as many for sake of argument). So by this reasoning, apache should have as many worms as IIS. But, as far as I can remember, there have only been two Apache worms. Neither of which btw were as crippling as any IIS worm. In fact, I was running multiple apache servers at the time of both of them and got neither one. What about Oracle? IIRC Oracle has a larger market share than sql server. Do we know of any RDBMS worms as devistating as slammer?
Microsoft still isn't taking security seriously. Although this virus requires user interaction, Microsoft shouldn't make it so easy to execute content. Hell, content can be executed just by looking at the preview pane in outlook. Check out the story over in developers. MS decided instead of fixing the url spoofing bug that phishers have been using since december, they are just going to not allow urls with an @ sign in them.
Then you've got your idiots over at security focus, such as Tim Mullen (who is a security consultant for MS btw) who believes security shouldn't be an issue for MS to worry about. It should be the end user who worries about it. It's no wonder they do not take security seriously when you've got people with views like that advising you.
Let's not forget the anti virus companies. Their lively hood is protecting people from virii. Not stoping them, protecting people from them. If we didn't have virii, then the anti virus companies would be out of business.
When you've got all this political bullshit swirling around the only one that loses is the end user. The one who bought their computer to enhance their life. To get onto the internet and reasearch car safety because their teenager is about to drive. Or the grandma who wants to recieve pictures from her grand children. Or the first time user that gets a virus within 15 minutes of plugging in their new computer, ensuring they will probably hate it from that point on.
The truth is, if you haven't done that little bit of effort (filter out executable data and have a mail virus scanner) then your business deserves to get this virus.
Every single network admin I know at one point realized the severity of the problem and got a measure of protection. CNN acts like it is the first email virus ever. Mail appliances with an anti virus on them can be obtained for under a thousand dollars. Just plug it into the network and configure it via web interface. For those without protection, yes it will be a huge headache. For those with protection, at most it will be a small bump on an average day. Maybe at most some bandwidth slowdown due to the extra mail traffic. Even then, mail from servers shown to be sending the virus can be blocked.
LMAO. Ok, let me get this straight. Instead of properly fixing the bug, they issue this extravagent work around. Is it really that hard to fix properly? I mean it's been over a month and people are very activly exploiting this problem. I remember MS promising us a couple of years ago their secured computing initiative (or what ever they called it). This doesn't look like much effort to me. Looks more like they don't want to fix the problem properly so they are using this work around of disabling it. Not saying @ urls are particularly useful, but it does make you what else they half ass internally that we never hear about.
I think www.sco.com as we know it will probably have traffic from this virus FOREVER. Virii don't go away. Hell, I still see hits from code red in my logs. How long ago was that? SCO is looking at the very least a week of MAJOR traffic, more likely at least a month. Then if somehow the virus dies down a bit, they will probably see a couple hundred megabytes of virus traffic a day at least.
Other spammers want to make money too. It needs to be all or nothing. Microsoft can not block others spam, yet still allow their own. That is the epitome of a hypocrit. I think spammers should be allowed to spam (within reason, no taking over open relay servers) because they are just assholes trying to make money. So is microsoft! What makes the two any different. Is it automactically not spam when it comes from big companies????
seeing at Hotmail sends me spam. Altough I know they don't consider it spam seeing as it's Microsoft. They also don't consider their pop ups "pop ups" persay...
That's what everyone forgets in all this... how much their products suck. If their products could hold their own based on quality and merit, we wouldn't have these issues because there wouldn't be a demand for office replacements. Because man, when's the last time you ran a critical server on Unixware...
http://www.theregister.co.uk/content/68/35052.html
Looks like a start to me (though it's not power management, it's a wifi driver. But Linux needs all the wifi driver writers it can get).
> We have RPM on Linux, and that should be okay to use here.
RPM's are awful. Distrubiting binaries instead of source in general is awful. The second you've linked it againest a different version of a lib it's usually unusable. The GNU build system takes care of that much better.
> An active-directory similar interface for distributing software packages.
Novell is in the process of porting ZenWorks which from what I understand can do this. Though I would find it hard to believe someone hasn't already written something to do this. It could be accomplished in a 200 line perl script.
>X crashes much too often
This I find VERY hard to believe. Xfree86 has definatly taken stability over features and has taken a lot of slack for it, but it definatly is stable. I'm running a RC (4.4.0 rc2) and don't even have problems with it. I ran 4.3.1 before that for a long time, never had problems with it. It's very agressivly compiled too. Some binary packagers have a way of turning a good program into an unstable binary (I'm looking in your direction Red Hat). Try compiling Xfree86 from source with just Make World.
>the time to restart X seem to match the time to restart a normal Windows XP computer
How are you restarting X? I can kill X with ctrl+alt+backspace and startx again in under 5 seconds.
OMG someone finally gets it!!! All this talk of there's no gui for this or that etc. etc. There have been plenty of GUI's for a looooong time. I remember back in the day using Mandrake (very old version) and that had a GUI package system. RPM cooker I think it was called. You just clicked the cooker icon, it updated list of the software that was available, and you either installed new software or updated old software. I didn't know anything about command line back then and was the average desktop user, and I used it just fine. I setup a home network with it just using the GUI. I know people who run linux servers and don't know any command line (usually run it through webmin or an included GUI). KDE has a GUI for freakin everything. There must be at least 10 GUI's just for installing RPM's. I think all this talk of "there needs to be more GUI's" is old, because there have been plenty of GUI's for years. I can't think of a single task that someone hasn't written a GUI frontend for, or at the very least an NCurses interface for.
I watch CNBC on a regular basis and let me tell you, I've never seen such a fickle group of people in my life. Their opinions of tech companies are very rarely based on sound technologies. They are based on press releases and analyst reports. Their most trusted analyst seems to be Gartner. When the hell has Gartner ever told us anything insightful??? But they treat Gartner as it has it's ear to the pulse of technology. Oh, and their second favorite is the Yankee group. Don't forget to mention how rarely they mention any of their parent companies when doing in depth reporting on them. CNBC is the epitome of capitalist piggery at it's worst. Yet, it's amazing how much affect their sheer ignorance and arrogance about knowledge of tech seems to affect the market. I've never seen anyone want to suck Intel's dick so badly in my life.
*waits to be modded down for being OT*
Slashdot Mirror
Some of your points are valid, some are not.
You mentioned installing Linux. How many average users do you know that installed their version of windows rather than bought it with it on there already? Most of those people wouldn't be able to install windows. The fact is, for so called "n00b" distro's, the installion process is in fact easier and quicker than windows. Some such as Lindows and Mandrake can be installed onto a new computer with just a couple of steps and nothing more than pressing next. PC's bundled with Linux and linux friendly hardware alieve this problem completely.
RPM's are shitty and are the cause of more problems than they are worth. RPM's should have been done away with long ago. I remember as a noob, same problems. Could never install RPM's because of dependencys. I think distro's should adopt a portage style package system. Compiling from source takes longer, but it takes care of idiotic linking issues that RPM's create.
I think most of Linux's entry level issues will be solved when OEM's start shipping it more included. The reason Linux on the desktop at the office is a great canidate is because it's installed for the workers and they've got someone to help them get past their entry level issues. Which are minor, but can be a big deal to new users.
They would make perfect real world sense because almost no Linux distro is compiled with the Intel cc. They are almost all compiled with GCC.
I may be the only one to feel this way... but haven't the completely unrelated SCO jokes gotten a little old? They are completely offtopic 99% of the time and most of them are just "Where do I send my 699 dollars".
I am willing to bet they've done _something_. For example, many times I tried to connect and my first try was immediatly stopped by differing error messages "This document contains no data" or "You don't have access to this resource". Then on a second attempt I am connect. Pretty much 9 times out of 10. Also, the initial connection to the main page might be slow; But once I'm on the page it seems fine. My belief is maybe they are purposly putting a latency on the first connection and then using keep-alive and allowing those conections more bandwidth. This would be similiar to a tar pit in that it's supposed to keep the client in a state of waiting as long as possible. The longer it waits, the less DoSing it can do. Just my two cents.
Groklaw has been slashdotted and I really want to read this so would someone be nice enough to post the article?
It's slashdot. Anything that questions the linux community is modded flamebait/troll.
I think it's really dumb so many companies are backing gnome when, it's really an inferior window manager. First of all, it's a package maintainers nightmare. Gnome is split up into eleventy twelve bazillion packages. While Kde really only needs QT, Kdelibs, and Kdebase to get up and running. QT is more mature, more stable, and simple a better toolkit than Gtk. Gtk is much more of a pain in the ass to program in than QT. Gnome was a project that came out of RedHat... 'nuff said ;-)
KDE has more to it. Things like KHtml and the kde regular expression library. A browser that's much more integrated into the wm than Galeon (have you tried the latest KDE rc? Konq opens faster than IE on windows). KWallet is awesome.
Not saying Gnome should go away. It's got some good individual packages. IIRC, Gnome is responsible for the XML related packages for linux. But why are so many supporting Gnome while ignoring KDE when it is such a great wm.
If you live your life like that you'll be using old software for a long long time.
The problem isn't autoconf. Autoconf is a hack to compensate for the defencies of Automake. And the basic idea of needing a scriping/macro language to figure out what packages are installed and how the system is laid out is going to have to be there as long there is no standard file(s) that contains all package and system information. I think another language besides m4 should be used (perl or python?). In fact, tools looking to replace autoconf do in fact use python (haven't seen any perl yet though). I think a lot of people fail to realize that most OSS comes in the form of ./configure && make && make install .deb, .rpm, .tgz binary packages are usually made with binaries that were compiled with the help of autoconf. Even gentoo's ebuild system uses autoconf.
Even those
The argument I hear the most, without a doubt "Windows gets more viruii because it's more popular". I call bullshit! I know it's bullshit because of Apache. Apache, by almost any web server survey, has at least as many servers as IIS (netcraft says between 2x and 3x, but let's say just as many for sake of argument). So by this reasoning, apache should have as many worms as IIS. But, as far as I can remember, there have only been two Apache worms. Neither of which btw were as crippling as any IIS worm. In fact, I was running multiple apache servers at the time of both of them and got neither one. What about Oracle? IIRC Oracle has a larger market share than sql server. Do we know of any RDBMS worms as devistating as slammer?
Microsoft still isn't taking security seriously. Although this virus requires user interaction, Microsoft shouldn't make it so easy to execute content. Hell, content can be executed just by looking at the preview pane in outlook. Check out the story over in developers. MS decided instead of fixing the url spoofing bug that phishers have been using since december, they are just going to not allow urls with an @ sign in them.
Then you've got your idiots over at security focus, such as Tim Mullen (who is a security consultant for MS btw) who believes security shouldn't be an issue for MS to worry about. It should be the end user who worries about it. It's no wonder they do not take security seriously when you've got people with views like that advising you.
Let's not forget the anti virus companies. Their lively hood is protecting people from virii. Not stoping them, protecting people from them. If we didn't have virii, then the anti virus companies would be out of business.
When you've got all this political bullshit swirling around the only one that loses is the end user. The one who bought their computer to enhance their life. To get onto the internet and reasearch car safety because their teenager is about to drive. Or the grandma who wants to recieve pictures from her grand children. Or the first time user that gets a virus within 15 minutes of plugging in their new computer, ensuring they will probably hate it from that point on.
The truth is, if you haven't done that little bit of effort (filter out executable data and have a mail virus scanner) then your business deserves to get this virus.
Every single network admin I know at one point realized the severity of the problem and got a measure of protection. CNN acts like it is the first email virus ever. Mail appliances with an anti virus on them can be obtained for under a thousand dollars. Just plug it into the network and configure it via web interface. For those without protection, yes it will be a huge headache. For those with protection, at most it will be a small bump on an average day. Maybe at most some bandwidth slowdown due to the extra mail traffic. Even then, mail from servers shown to be sending the virus can be blocked.
LMAO. Ok, let me get this straight. Instead of properly fixing the bug, they issue this extravagent work around. Is it really that hard to fix properly? I mean it's been over a month and people are very activly exploiting this problem. I remember MS promising us a couple of years ago their secured computing initiative (or what ever they called it). This doesn't look like much effort to me. Looks more like they don't want to fix the problem properly so they are using this work around of disabling it. Not saying @ urls are particularly useful, but it does make you what else they half ass internally that we never hear about.
I mean who's to say one of their other eleventy tweleve bagillion enemies didn't write it?
I think www.sco.com as we know it will probably have traffic from this virus FOREVER. Virii don't go away. Hell, I still see hits from code red in my logs. How long ago was that? SCO is looking at the very least a week of MAJOR traffic, more likely at least a month. Then if somehow the virus dies down a bit, they will probably see a couple hundred megabytes of virus traffic a day at least.
Other spammers want to make money too. It needs to be all or nothing. Microsoft can not block others spam, yet still allow their own. That is the epitome of a hypocrit. I think spammers should be allowed to spam (within reason, no taking over open relay servers) because they are just assholes trying to make money. So is microsoft! What makes the two any different. Is it automactically not spam when it comes from big companies????
seeing at Hotmail sends me spam. Altough I know they don't consider it spam seeing as it's Microsoft. They also don't consider their pop ups "pop ups" persay...
That's what everyone forgets in all this... how much their products suck. If their products could hold their own based on quality and merit, we wouldn't have these issues because there wouldn't be a demand for office replacements. Because man, when's the last time you ran a critical server on Unixware...
or the dildos, or the anal beads, or the chocolate flavored condoms....
So now she knows at what velocity it'll hit her in the face.
http://www.theregister.co.uk/content/68/35052.html
Looks like a start to me (though it's not power management, it's a wifi driver. But Linux needs all the wifi driver writers it can get).
> We have RPM on Linux, and that should be okay to use here.
RPM's are awful. Distrubiting binaries instead of source in general is awful. The second you've linked it againest a different version of a lib it's usually unusable. The GNU build system takes care of that much better.
> An active-directory similar interface for distributing software packages.
Novell is in the process of porting ZenWorks which from what I understand can do this. Though I would find it hard to believe someone hasn't already written something to do this. It could be accomplished in a 200 line perl script.
>X crashes much too often
This I find VERY hard to believe. Xfree86 has definatly taken stability over features and has taken a lot of slack for it, but it definatly is stable. I'm running a RC (4.4.0 rc2) and don't even have problems with it. I ran 4.3.1 before that for a long time, never had problems with it. It's very agressivly compiled too. Some binary packagers have a way of turning a good program into an unstable binary (I'm looking in your direction Red Hat). Try compiling Xfree86 from source with just Make World.
>the time to restart X seem to match the time to restart a normal Windows XP computer
How are you restarting X? I can kill X with ctrl+alt+backspace and startx again in under 5 seconds.
OMG someone finally gets it!!! All this talk of there's no gui for this or that etc. etc. There have been plenty of GUI's for a looooong time. I remember back in the day using Mandrake (very old version) and that had a GUI package system. RPM cooker I think it was called. You just clicked the cooker icon, it updated list of the software that was available, and you either installed new software or updated old software. I didn't know anything about command line back then and was the average desktop user, and I used it just fine. I setup a home network with it just using the GUI. I know people who run linux servers and don't know any command line (usually run it through webmin or an included GUI). KDE has a GUI for freakin everything. There must be at least 10 GUI's just for installing RPM's. I think all this talk of "there needs to be more GUI's" is old, because there have been plenty of GUI's for years. I can't think of a single task that someone hasn't written a GUI frontend for, or at the very least an NCurses interface for.
I've got 5 words for them... Shut you're fucking face unclefucker!!! =) (yeah, I know, uncle fucker is two words. But it should be one!)
I watch CNBC on a regular basis and let me tell you, I've never seen such a fickle group of people in my life. Their opinions of tech companies are very rarely based on sound technologies. They are based on press releases and analyst reports. Their most trusted analyst seems to be Gartner. When the hell has Gartner ever told us anything insightful??? But they treat Gartner as it has it's ear to the pulse of technology. Oh, and their second favorite is the Yankee group. Don't forget to mention how rarely they mention any of their parent companies when doing in depth reporting on them. CNBC is the epitome of capitalist piggery at it's worst. Yet, it's amazing how much affect their sheer ignorance and arrogance about knowledge of tech seems to affect the market. I've never seen anyone want to suck Intel's dick so badly in my life. *waits to be modded down for being OT*
SCO accusing others of slander? That's like Jenna Jameson accusing the local librarian of having a loose puss.