Here's the quote you're looking for (circa January 1992 by Andy Tanebaum, "LINUX is obsolete" in comp.os.minix):
Of course 5 years from now that will be different, but 5 years from now everyone will be running free GNU on their 200 MIPS, 64M SPARCstation-5.
Part of a huge flamewar between Linus, AST and a number of other people on microkernel vs monolithic kernel design. Here's the entire original thread if you're looking for some good Sunday reading.
You mean the Xine engine? That's what I have it set to now. While the gaps are much smaller than they used to be with the previous version of amaroK, they're still present (albeit very small).
So let's say you have just installed an OS on a computer, and you want 3d acceleration with your Radeon 9600 to work.
Let's say we tried this out with a Radeon 9200 instead. Oh, we're already done because that card is supported by X.org.
This is the reason that we prefer open source drivers to binary drivers, they can be maintained properly when they need to be, rather than waiting on the vendor. Specs have been released for the Radeons 9200 and under, thus they're well supported. NVidia is very good about keeping their Linux drivers up to date, thus you won't hear many complaints about them, but ATI is a different matter.
I agree mostly with the parent poster, in terms of features, the only thing that XMMS holds over Amarok is gapless playback. While the latest version of Amarok got much better in this respect, it's still not quite as crisp as XMMS. Well, that and Amarok takes up a lot more memory and CPU time, but yeah, it's a very nice player otherwise.
I keep trying Linux, and keep moving back to windows. Each time I go Linux I stay there for longer and longer. Until I do stupid things and Linux wont boot because I tried to re-compile the kernal without reading any docs.
In the future, try adding the new kernel as an additional boot option, rather than replacing the old one. Trivially simple to do with LILO and I would presume the same with Grub, though I've never used Grub myself. you should have System.map-$VERSION, config-$VERSION and kernel-$VERSION in your/boot directory, open up your LILO config file (usually in/etc/lilo.conf), copy and paste the block of text that you're using for your current kernel, and replace the version numbers, label it something that you'll recognize. Save the file and run lilo as root, it'll update the MBR and you'll see a new option the next time you boot. If the new kernel doesn't boot, you can just pick the old kernel and try again.
That's not an accurate assessment. According to an NPD Online - Worldwide Survey 97.6% of Internet enabled PCs support Flash 5.
Those numbers seem very high to me, I'd also consider where you're reading them (the Macromedia site...) for the obvious bias. Windows 98 had Flash as an optional install, it wasn't installed by default IIRC. Windows 2000 does not include Flash on the CD, or, if it does, it's not listed in "Add/Remove Windows Components" anymore, and is not installed by default. I have no idea about Windows XP or ME as I wouldn't touch them with a 10 foot clown pole, and I don't think Flash existed in time for Windows 95. It is very possible that XP does not include Flash by default however as I've seen many XP machines without it.
On top of that, even if they have Flash installed, a lot of people will disable Flash because it's so prevalent in (annoying) ads. You certainly won't find Flash on any of my computers.
Yes there are many incompatibilities, so the only reason you haven't encountered any incompatibilities, is because you haven't tested your application on every possible browser and operating system, which you have to do in order to be sure your AJAX application will run across any platforms you want to support.
While not exhaustive, my use of AJAX works properly in Konqueror, Firefox, Opera, Netscape, Safari, and IE across Linux, OS X, Windows, and even a little bit of *BSD. That covers the vast majority of browsers, and it falls back on other methods for browsers without XMLHttpRequest. It didn't require any tinkering beyond the initial workaround for IE either.
You couldn't be more mistaken about the XMLHTTPRequest implementation in Internet Explorer being non-standard. In case you didn't know, XMLHTTPRequest was invented by Microsoft, and Internet Explorer was the first browser to support it. Microsoft's implementation of XMLHTTPRequest defines the interface, and all other browsers are implementations of Microsoft's specification. Better do your homework next time, kiddo.
None of that is news to me, however, if you'll read my original post, you'll understand what I'm talking about. The code snippet I linked to details how to create the initial XMLHttpRequest object in a portable manner, because as per usual, IE does it's own thing, while every other browser has the same interface. After you've got the initial object, yes, they all have the same interface, but not before that. As always, I don't design for IE, I write standards compliant code, and test in Konqueror and Firefox first. After that, I test in IE to make sure it doesn't mangle it too badly, but IE just flat out doesn't support a lot of things (such as position:fixed), so it's a matter of hoping it degrades gracefully in IE.
AJAX requires you to test against all browsers, because all browsers are subtly and not-so-subtly different. If you're not doing all that testing, then you have no way of knowing whether or not your AJAX application will really work on other people's browsers, and it probably won't.
You can say that about any web technology if you implement it for IE first. That shouldn't surprise anyone here. However, that's not true in the vast majority of cases when you disregard IE (to start).
Honestly, that's why I moved to Flash. It just works on all the browsers, and I never have to retrofit changes to keep up compatibility.
If they have the Flash plugin installed that is. I don't have any numbers, but I'd wager that the number of people with Flash (perhaps 30%?) is much much smaller than the number of people with an AJAXable browser (about 98%?). People are lazy, they don't want to go install a plugin to view your site. Why do you think IE still has so many users? How many corporate desktops do you think have Flash installed by default? If you ask me, Flash is a terrible way to try and reach a large audience.
"AJAX" will work with Internet Explorer, Gecko-based browsers, and Apple's Safari, and nothing else, because these are the only browsers to have included implementations of the XMLHTTPRequest object yet.
Quite untrue. Konqueror supports AJAX, I'm posting from it right now, and my Online RPG uses it in a few places. Opera also supports AJAX. That leaves us with what... text based or PDA browsers? Most of those don't even support Javascript, good luck getting Flash or something else similar working on it. You never did mention what you would prefer over AJAX, what would you prefer over AJAX anyways? I presume that whatever you would prefer would magically work on cell phones, PDAs and text based browsers too, as you seem to think that that's a major issue for AJAX.
Why does OWE degrade in Firefox? Doesn't "AJAX" work "off the shelf" in all major browsers? Well, the problem is that
each of these browsers has their own incompatible implementation of XMLHTTPRequest.
If they have, I haven't encountered it. Besides IE doing it's own thing like usual, every other browser has the same implementation. Unusually enough, in this case, working around IE is pretty simple.
Now you have to program a second, stripped down, clunky version just for that contingency.
That can be said about the majority of web technologies, if you want to support IE, you get to jump through lots of hoops. If you don't care about IE, you can write standards compliant code and expect it to work and look the same in Konqueror, Firefox, Opera, Safari, etc. As always with web development, IE is the problem, working around it is where you will spend/waste the most time.
Having said all of that, like any other technology, you should use AJAX to enhance the experience, not have everything depend on it. Your web site/application/whatever should still be accessable without all of the handy dandy technology, regardless of what it is. That's much harder if you've used Flash or anything else that depends on a plugin for everything, because it means you need to totally rewrite code.
TCP has an overhead of approximately 11%, so the 10:1 ratio seems about right. Also keep in mind that if you saturate the upstream on a typical cable connection, you'll really kill it in terms of response, whereas the same generally isn't true if you saturate your downstream.
Results 1 - 10 of about 6,990,000 for linux kde. (0.06 seconds)
Results 1 - 10 of about 1,850,000 for linux kde bsd. (0.22 seconds)
Results 1 - 10 of about 1,390,000 for kde bsd. (0.15 seconds)
Results 1 - 10 of about 5,410,000 for linux gnome. (0.05 seconds)
Results 1 - 10 of about 1,260,000 for linux gnome bsd. (0.14 seconds)
Results 1 - 10 of about 1,400,000 for gnome bsd. (0.03 seconds)
Looks like Gnome edges out KDE on BSD ever so slightly. Then again, we all know BSD is dying, so perhaps not so relevant?
Results 1 - 10 of about 13,500,000 for kde. (0.03 seconds)
Results 1 - 10 of about 11,400,000 for gnome [definition]. (0.05 seconds)
It looks like KDE is at least talked about more than Gnome. Not to mention that Gnome probably turned up a number of results for other sorts of Gnomes (Lawn Gnomes, D&D Gnomes, etc), whereas I'm not aware of any other meaning for KDE.
Have you ever tried to get an X server, accelerated 3D, and a framebuffer console to get along on the same machine? It's ugly.
Maybe it's just me, but my ThinkPad T40 has all of the above, and it was pretty painless to get working. If using Gentoo, just make sure you load radeonfb and not vesafb. 3D acceleration was a zero configuration matter, as the Radeon 7500 is already supported natively by X.org.
My Desktop machine also has all of the mentioned features using a GeForce 3. The framebuffer console "just worked", and so did the 3D acceleration after I installed the NVidia drivers.
Disclaimer: I've not tried the binary ATI drivers because I don't have the necessary hardware to do so. I have no idea how well they work. The same goes for multiple monitors.
Essentially, that the file does not exist, or cannot be read, as is explained immediately after in the error message.
2) Is that the path of the source, or the destination it can't find?
It can't find the source file, this should be obvious as that is the file it references in the error message. Also, since not being able to find the source is quite clearly a fatal error, so it doesn't bother checking the destination.
In any case, in my problem that message was utterly useless, because the "path" was "~\windows_drivers\something.ini" and the path I was expecting to see was (gasp!) "~\windows_drivers\something.ini". The problem is that "~" means something different in a root terminal than it does in every other terminal... interestingly, Apple's fixed that problem, but Ubuntu hasn't yet.
I'm pretty sure shell expansion can be configured to your liking, but using the first example again in my copy of bash and cp:
neil@t40-n ~ $ cp ~/abc ~/def
cp: cannot stat `/home/neil/abc': No such file or directory
Clearly in this case, shell expansion is on and the varying meaning of ~ is not a problem.
(My latest install of FC4 had my notebook's speedstep running at ~600Mhz even when plugged into AC, until I manually tweaked some files.)
I'm not using FC4, however I am using cyudynd to manage Speedstep on my laptop, and it does exactly this (because I want it to). What you probably didn't notice is that if you actually need the extra processor power, it'll immediately clock back up until you don't need it anymore. I've played with (and benchmarked) it a fair bit and there is essentially no difference in response with Speedstep enabled (and thus jumping between 600 Mhz and 1.3 Ghz accordingly) compared to just locking the processor at 1.3 Ghz. I did however notice a huge difference in the heat produced and the battery life of the laptop, with Speedstep enabled, the fan almost never needs to turn on, and the laptop runs much cooler.
3) What to do for common errors. When I tried the above, I got some error like "cp: Failed, file does not exist." Well, I know now in retrospect that it's because I was telling it to look in the wrong desktop folder, but the How-To didn't have any explaination of that error. (And no, I don't know that "cp" means "file copy." And again, if you're looking for reasons why people hate Linux,
how about the obvious: If "cp" runs into an error, why doesn't it tell you WHICH FILE has the error? I mean, duh!)
Maybe it's just my version of cp, but...
neil@t40-n ~ $ cp abc def
cp: cannot stat `abc': No such file or directory
What Linux Needs (tm) to really get established at home (in my humble opinion) is a complete end to end installer for apps and drivers. End to end means that you choose an app to install and the installer also installs any dependent libraries WITHOUT asking you where they are on the internet, and compiles the dependencies from source if it isn't available from your distro already compiled, and it handles the architectural switches (x86 vs. amd64 for example), and it ties the new app into the Windows Manager you are using (such as creating the icon to run the app from the WM menu).
Another Thing That Wouldn't Hurt (tm) is a central repository for links to non-OSS packages, especially drivers. Since most distros don't include proprietary drivers, sometimes it is tough to find them. My ATI graphics card is one such example, my Broadcom wireless networking card is another.
You said that you tried Gentoo already, but it seems that you didn't look too carefully at it, as it really does all of these things already. Portage is the best thing since sliced bread, although it's not really alone in that respect. I find the biggest mistake that new users make with regards to Linux is that they totally ignore their package manager, then get frustrated because they want the source tarballs to do all the things that the package manager already does. (I did it too when I was new to Linux...) Of course the blame lies in the fact that there is no package manager at all in Windows, and too many people equate Windows with computing, thinking that everything must be like it. I don't think there are any package managers that don't already do automagic dependency checking and resolving, although if you snag random rpms off the net and try and install them you may run into problems (which is why you always check your package manager first!).
As to non-free items, both the ATI and NVidia binary drivers are one emerge away (emerge ati-drivers or emerge nvidia-glx, respectively). You can specify to use win32 codecs (set win32codecs in your USE flags) for Mplayer, Xine and most every other media player on Linux, and Portage will snag those automagically too. I really can't say enough good things about Portage, I've never been happier.
Slashdot Mirror
Read my post again and you'll notice that I didn't say any of that, you merely assumed it.
He argued that it's wrong to download things because it's illegal, I pointed out the problem with that argument.
Illegal does not imply morally wrong. Not all laws are justified or "correct".
Part of a huge flamewar between Linus, AST and a number of other people on microkernel vs monolithic kernel design. Here's the entire original thread if you're looking for some good Sunday reading.
You mean the Xine engine? That's what I have it set to now. While the gaps are much smaller than they used to be with the previous version of amaroK, they're still present (albeit very small).
I may only speak for myself here, but the only DRM that's OK with me is no DRM at all.
Let's say we tried this out with a Radeon 9200 instead. Oh, we're already done because that card is supported by X.org.
This is the reason that we prefer open source drivers to binary drivers, they can be maintained properly when they need to be, rather than waiting on the vendor. Specs have been released for the Radeons 9200 and under, thus they're well supported. NVidia is very good about keeping their Linux drivers up to date, thus you won't hear many complaints about them, but ATI is a different matter.
I agree mostly with the parent poster, in terms of features, the only thing that XMMS holds over Amarok is gapless playback. While the latest version of Amarok got much better in this respect, it's still not quite as crisp as XMMS. Well, that and Amarok takes up a lot more memory and CPU time, but yeah, it's a very nice player otherwise.
In the future, try adding the new kernel as an additional boot option, rather than replacing the old one. Trivially simple to do with LILO and I would presume the same with Grub, though I've never used Grub myself. you should have System.map-$VERSION, config-$VERSION and kernel-$VERSION in your /boot directory, open up your LILO config file (usually in /etc/lilo.conf), copy and paste the block of text that you're using for your current kernel, and replace the version numbers, label it something that you'll recognize. Save the file and run lilo as root, it'll update the MBR and you'll see a new option the next time you boot. If the new kernel doesn't boot, you can just pick the old kernel and try again.
The Radeon DRI drivers support up to and including the Radeon 9200/9250. There is experimental 3D acceleration support for the Radeons above this.
Those numbers seem very high to me, I'd also consider where you're reading them (the Macromedia site...) for the obvious bias. Windows 98 had Flash as an optional install, it wasn't installed by default IIRC. Windows 2000 does not include Flash on the CD, or, if it does, it's not listed in "Add/Remove Windows Components" anymore, and is not installed by default. I have no idea about Windows XP or ME as I wouldn't touch them with a 10 foot clown pole, and I don't think Flash existed in time for Windows 95. It is very possible that XP does not include Flash by default however as I've seen many XP machines without it.
On top of that, even if they have Flash installed, a lot of people will disable Flash because it's so prevalent in (annoying) ads. You certainly won't find Flash on any of my computers.
While not exhaustive, my use of AJAX works properly in Konqueror, Firefox, Opera, Netscape, Safari, and IE across Linux, OS X, Windows, and even a little bit of *BSD. That covers the vast majority of browsers, and it falls back on other methods for browsers without XMLHttpRequest. It didn't require any tinkering beyond the initial workaround for IE either.
None of that is news to me, however, if you'll read my original post, you'll understand what I'm talking about. The code snippet I linked to details how to create the initial XMLHttpRequest object in a portable manner, because as per usual, IE does it's own thing, while every other browser has the same interface. After you've got the initial object, yes, they all have the same interface, but not before that. As always, I don't design for IE, I write standards compliant code, and test in Konqueror and Firefox first. After that, I test in IE to make sure it doesn't mangle it too badly, but IE just flat out doesn't support a lot of things (such as position:fixed), so it's a matter of hoping it degrades gracefully in IE.
You can say that about any web technology if you implement it for IE first. That shouldn't surprise anyone here. However, that's not true in the vast majority of cases when you disregard IE (to start).
If they have the Flash plugin installed that is. I don't have any numbers, but I'd wager that the number of people with Flash (perhaps 30%?) is much much smaller than the number of people with an AJAXable browser (about 98%?). People are lazy, they don't want to go install a plugin to view your site. Why do you think IE still has so many users? How many corporate desktops do you think have Flash installed by default? If you ask me, Flash is a terrible way to try and reach a large audience.
Quite untrue. Konqueror supports AJAX, I'm posting from it right now, and my Online RPG uses it in a few places. Opera also supports AJAX. That leaves us with what... text based or PDA browsers? Most of those don't even support Javascript, good luck getting Flash or something else similar working on it. You never did mention what you would prefer over AJAX, what would you prefer over AJAX anyways? I presume that whatever you would prefer would magically work on cell phones, PDAs and text based browsers too, as you seem to think that that's a major issue for AJAX.
If they have, I haven't encountered it. Besides IE doing it's own thing like usual, every other browser has the same implementation. Unusually enough, in this case, working around IE is pretty simple.
That can be said about the majority of web technologies, if you want to support IE, you get to jump through lots of hoops. If you don't care about IE, you can write standards compliant code and expect it to work and look the same in Konqueror, Firefox, Opera, Safari, etc. As always with web development, IE is the problem, working around it is where you will spend/waste the most time.
Having said all of that, like any other technology, you should use AJAX to enhance the experience, not have everything depend on it. Your web site/application/whatever should still be accessable without all of the handy dandy technology, regardless of what it is. That's much harder if you've used Flash or anything else that depends on a plugin for everything, because it means you need to totally rewrite code.
PHP: Hypertext Preprocessor
TCP has an overhead of approximately 11%, so the 10:1 ratio seems about right. Also keep in mind that if you saturate the upstream on a typical cable connection, you'll really kill it in terms of response, whereas the same generally isn't true if you saturate your downstream.
Results 1 - 10 of about 6,990,000 for linux kde. (0.06 seconds)
Results 1 - 10 of about 1,850,000 for linux kde bsd. (0.22 seconds)
Results 1 - 10 of about 1,390,000 for kde bsd. (0.15 seconds)
Results 1 - 10 of about 5,410,000 for linux gnome. (0.05 seconds)
Results 1 - 10 of about 1,260,000 for linux gnome bsd. (0.14 seconds)
Results 1 - 10 of about 1,400,000 for gnome bsd. (0.03 seconds)
Looks like Gnome edges out KDE on BSD ever so slightly. Then again, we all know BSD is dying, so perhaps not so relevant?
Results 1 - 10 of about 13,500,000 for kde. (0.03 seconds)
Results 1 - 10 of about 11,400,000 for gnome [definition]. (0.05 seconds)
It looks like KDE is at least talked about more than Gnome. Not to mention that Gnome probably turned up a number of results for other sorts of Gnomes (Lawn Gnomes, D&D Gnomes, etc), whereas I'm not aware of any other meaning for KDE.
As a side note, I don't think I've ever seen Gnome ahead of KDE in a poll before.
Same here. 10% seems a little high for either of those populations however...
Maybe it's just me, but my ThinkPad T40 has all of the above, and it was pretty painless to get working. If using Gentoo, just make sure you load radeonfb and not vesafb. 3D acceleration was a zero configuration matter, as the Radeon 7500 is already supported natively by X.org.
My Desktop machine also has all of the mentioned features using a GeForce 3. The framebuffer console "just worked", and so did the 3D acceleration after I installed the NVidia drivers.
Disclaimer: I've not tried the binary ATI drivers because I don't have the necessary hardware to do so. I have no idea how well they work. The same goes for multiple monitors.
Essentially, that the file does not exist, or cannot be read, as is explained immediately after in the error message.
It can't find the source file, this should be obvious as that is the file it references in the error message. Also, since not being able to find the source is quite clearly a fatal error, so it doesn't bother checking the destination.
I'm pretty sure shell expansion can be configured to your liking, but using the first example again in my copy of bash and cp:
neil@t40-n ~ $ cp ~/abc ~/def
cp: cannot stat `/home/neil/abc': No such file or directory
Clearly in this case, shell expansion is on and the varying meaning of ~ is not a problem.
I'm not using FC4, however I am using cyudynd to manage Speedstep on my laptop, and it does exactly this (because I want it to). What you probably didn't notice is that if you actually need the extra processor power, it'll immediately clock back up until you don't need it anymore. I've played with (and benchmarked) it a fair bit and there is essentially no difference in response with Speedstep enabled (and thus jumping between 600 Mhz and 1.3 Ghz accordingly) compared to just locking the processor at 1.3 Ghz. I did however notice a huge difference in the heat produced and the battery life of the laptop, with Speedstep enabled, the fan almost never needs to turn on, and the laptop runs much cooler.
Maybe it's just my version of cp, but...
neil@t40-n ~ $ cp abc def
cp: cannot stat `abc': No such file or directory
You said that you tried Gentoo already, but it seems that you didn't look too carefully at it, as it really does all of these things already. Portage is the best thing since sliced bread, although it's not really alone in that respect. I find the biggest mistake that new users make with regards to Linux is that they totally ignore their package manager, then get frustrated because they want the source tarballs to do all the things that the package manager already does. (I did it too when I was new to Linux...) Of course the blame lies in the fact that there is no package manager at all in Windows, and too many people equate Windows with computing, thinking that everything must be like it. I don't think there are any package managers that don't already do automagic dependency checking and resolving, although if you snag random rpms off the net and try and install them you may run into problems (which is why you always check your package manager first!).
As to non-free items, both the ATI and NVidia binary drivers are one emerge away (emerge ati-drivers or emerge nvidia-glx, respectively). You can specify to use win32 codecs (set win32codecs in your USE flags) for Mplayer, Xine and most every other media player on Linux, and Portage will snag those automagically too. I really can't say enough good things about Portage, I've never been happier.
Microsoft makes standards?