Sometimes you have to live in the real world, and you discover that not everything follows a single paradigm. Languages that follow a single paradigm have serious drawbacks. Java is one. There's a reason why Java isn't used for systems programming. There's a reason why Corel has yet to finish it's Java office suite.
An int is four bytes on my CPU. Why should I have the overhead of an object wrapped around it? Why do I need runtime polymorphism on ints? For OO educational purposes, it makes sense to teach that an int is an object. But often in the real world it's far better to make an int simply four bytes in memory.
Rule of thumb: if polymorphism doesn't make sense for an object, maybe it shouldn't be an object. What can you possibly derive from a bool that wouldn't still be a primitive bool?
I've always like the C++ iostreams way of printing. It's not pure OO, but it's intuitive. It just needs another set of operators.
I think where Java gets it wrong, and why "System.out.println()" looks so silly to you, is that Java students are taught that everything is an object. But not everything is an object, especially when you're printing.
It's not the easiest thing in the world. The typical "configure; make; make install" won't do it. If you're not set on your distro, you might try Gentoo or one of the BSD's and build from source. Or if you are set on it, check your distro's sources for XFree. Sometimes a build script or.spec file is included that you can tweak for an optimized build.
I have Win2K and FreeBSD/KDE dual booting on my workstation here at work. In my experience, it is KDE that is snappy and responsive while Win2K is the fat slug.
Suggestions: rebuild XFree86, Qt and KDE to be optimized for your compiler. Odds are you distro has them built for a generic i386. Make sure you have your system configured correctly.
but why is there KDE and Gnome? Why are the two not one.
Because freedom is choice and the control over your own destiny. Open Source and Free Software is not about monolithic environments that you must use. It's not about some elite group deciding what is best for you.
This freedom allows you to choose between KDE, GNOME, XFCE, Windowmaker, Enightenment, Blackbox or whatever. It allows you to choose between Redhat, SuSE, Gentoo, Manduck or Slackware. It allows you to choose between Linux, FreeBSD, OpenBSD and NetBSD. Choose between Emacs, XEmacs, Vim, nvi, elvis, joe, jed, kwrite, kate, gedit and dozens more.
What is so fundamentally different between Gnome and KDE that doesn't allow them to be merged into one project?
1) Politics. GNOME is Free Software and KDE is Open Source. Personally, I can't tell the difference between the two, but independent experimentation has shown that if you put a hardcore FSF/GNU person in the same room as an Open Source advocate, you can actually warp the curvature of time/space. And besides, GNOME is a GNU project and KDE is not, and for some people that makes all the difference in the world.
2) Language. GNOME is C and KDE is C++. Both have bindings to other languages, but for all practical purposes, one is C and the other C++.
3) Toolkits. First off, you have Qt versus GTK+. Until you can merge these two into one toolkit, you will never be able to merge KDE and GNOME. Both have different APIs, different OO paradigms, and different scopes.
IMHO that's all that is required to finally get a solid Desktop presence for linux.
If it is required for Linux to have a monolithic core of elites deciding what software must be used, before it can get on the desktop, then perhaps the desktop is not the right goal.
But no matter to me, since I use FreeBSD instead of Linux. There's choice for you!
A quick check of WIDI shows that Debian gets a 26%. That's a far cry from the 48% this survey shows. As for the OSE study, Debian was one of their sample pools to start with, so of course it's going to be skewed in that direction.
Here's my minority report from off the deep end. It sounds like to me that you will have someone maintain the boxes, and that all the user will ever see is the desktop. Fine. Then try FreeBSD.
FreeBSD is very easy to administer and has all the software Linux has. Stability and security is your prime concern in a public environment like this, and FreeBSD holds its own here. Only a few Linux distros can compare in this area (and the for-the-masses distros aren't them).
Every sampling method is skewed, even door-to-door canvassing. But Geez Louise! Trying to draw meaningful conclusions from a 100% self-selected sample is the epitome of anti-statistics.
Next you're going to argue for the robustness of the Slashdot Poll, aren't you?
This survey consisted entirely of self-selected participants. The surveyors actually seemed to boast that they didn't select anyone themselves. This study is bogus and proves nothing.
The staggering support for Debian (48%!) only proves that Debian developers are more successful in recruiting other Debian developers to participate in bogus online survey's. This isn't a dig against Debian, it's a dig against the silly methodology of this study.
This survey, despite its seemingly thoroughness, is no more valid than the weekly Slashdot Poll. The only conclusion that can be drawn from this survey is that Debian, GNOME and vi users are more likely to participate in a self-selected survey than Redhat, SuSE, Mandrake, KDE and emacs users.
Somehow I have a hard time believing that a bunch of developers who say "we won't follow any sensible UI design until we get mandatory guidelines" are the sort of people that would follow the guidelines to begin with.
What GNOME really needs is a release manager with the cojones to kick out anything that doesn't follow the standards.
I'm trying to remember Scott McNealy's exact quote at LWCE. Something like "Buy your Linux PC at Walmart instead of Dell. You'll get the same service."
But if the hardcore "compile or die" guys impose their view of computing on it, open source will fail fairly quickly
Who said anything about imposing our choice on you? We don't feel the need for a single unified free unix-like operating system. We don't think that only one distribution should come out on top. When we hear Linus say "world domination" we recognize that it is humor and not a goal.
I am perfectly happy with you using Mandrake just as long as you let me use Gentoo and FreeBSD.
I never used Mandrake long enough to get to this level of frustration. I left Mandrake a long time before that could happen. I suggest you do the same.
But I do agree somewhat with your basic premise. Mandrake seems to operate under the premise that "easy to use" means "no intermediate or advanced users allowed". Oh, advanced Mandrake users are fine, but if you have any knowledge of Unix, Posix or other distros, you're left outside in the cold.
The best thing about Corel LinuxOS was that it was small. Really. It asked you about three or four questions, then installed a very minimal default. You got a KDE desktop and that was it. On that desktop was an icon that allowed you to download a million other packages, but you didn't do that during the install.
I watched a greenhorn newbie install SuSE once. It was painful. He didn't have enough room for a complete full install, so he spent hours choosing between jed, joe, jove, emacs, xemacs, nvi, vim, elvis, pico and and a million other text editors. A newbie shouldn't have to choose between 7000 packages during the install. That is ridiculous.
An easy to use Linux installer will install just the basics necessary to get to a desktop. Even that minimal set of software is still going to give the user ten times the functionality of the full Windows install.
No, no, no! The United States is the Great Satan. We must attack it always with complete disregard to the truth. Automobiles are only driven in the US, polluted rivers only occur in the US, and only US oil tankers leak. This is the result of capitalism, which only exists in the US.
Back in my day if you went to college you had to take humanities and science classes. Stuff like foreign languages, composition, biology and philosophy were requirements. Even the engineering students. To make it fair, the liberal arts students had to take calculus as well.
But you tell that to youngsters nowadays and they won't believe you!
We give our code away to everyone, rich or poor, short or tall, stupid or smart. We don't attach strings to it. If you don't like Bill's copy of the code, download your own.
As a longtime BSD advocate, I must partly disagree. SELinux must be under the GPL, because it's based on GPLd software. The NSA can't arbitrarily change the license. As long as the choose to create derivative products from GPLd code, they must use the GPL. Besides which, public domain is incompatible with the GPL.
p.s. On the other hand, the NSA getting involved in a hardened BSD OS would be awesome.
but mostly talks about the much more benign stuff lumped under "direct marketing," like reminder updates from stores you cleared to send it to you.
"Hello, you are receiving this message because you selected to receive such messages on our website, one of our competitor's websites, or a completely unrelated website. If you do not wish to receive further messages of this type, please verify the validity of your email address by visiting the following address with a cookie-enabled browser. By removing your address from our list, you indicate your wish to receive similar messages of this type.
Every change like that can have hundreds of ramifications
Well duh! That's why they didn't make a release ninety minutes later!
If the fix had remained on the developers desk, no one could have possibly tested it. But by committing it to CVS, the testing can begin.
As a professional software developer, I expect I could have done a moderate amount of unit testing before I committed a fix like this. But there's no way in hell anyone is going to be able to do any integration or systems test until I commit it to the code base.
Code review and QA testing occur in the CVS tree. Neither can happen until the code is committed. Or are you recommending that KDE start using a private, restricted and closed development environment?
Slashdot Mirror
I don't get it. I must be missing something obvious. To me it's just annoying. But I don't understand mouse gestures either, so maybe it's just me.
Sometimes you have to live in the real world, and you discover that not everything follows a single paradigm. Languages that follow a single paradigm have serious drawbacks. Java is one. There's a reason why Java isn't used for systems programming. There's a reason why Corel has yet to finish it's Java office suite.
An int is four bytes on my CPU. Why should I have the overhead of an object wrapped around it? Why do I need runtime polymorphism on ints? For OO educational purposes, it makes sense to teach that an int is an object. But often in the real world it's far better to make an int simply four bytes in memory.
Rule of thumb: if polymorphism doesn't make sense for an object, maybe it shouldn't be an object. What can you possibly derive from a bool that wouldn't still be a primitive bool?
I've always like the C++ iostreams way of printing. It's not pure OO, but it's intuitive. It just needs another set of operators.
I think where Java gets it wrong, and why "System.out.println()" looks so silly to you, is that Java students are taught that everything is an object. But not everything is an object, especially when you're printing.
I have no idea how to rebuild XFree
.spec file is included that you can tweak for an optimized build.
It's not the easiest thing in the world. The typical "configure; make; make install" won't do it. If you're not set on your distro, you might try Gentoo or one of the BSD's and build from source. Or if you are set on it, check your distro's sources for XFree. Sometimes a build script or
I have Win2K and FreeBSD/KDE dual booting on my workstation here at work. In my experience, it is KDE that is snappy and responsive while Win2K is the fat slug.
Suggestions: rebuild XFree86, Qt and KDE to be optimized for your compiler. Odds are you distro has them built for a generic i386. Make sure you have your system configured correctly.
but why is there KDE and Gnome? Why are the two not one.
Because freedom is choice and the control over your own destiny. Open Source and Free Software is not about monolithic environments that you must use. It's not about some elite group deciding what is best for you.
This freedom allows you to choose between KDE, GNOME, XFCE, Windowmaker, Enightenment, Blackbox or whatever. It allows you to choose between Redhat, SuSE, Gentoo, Manduck or Slackware. It allows you to choose between Linux, FreeBSD, OpenBSD and NetBSD. Choose between Emacs, XEmacs, Vim, nvi, elvis, joe, jed, kwrite, kate, gedit and dozens more.
What is so fundamentally different between Gnome and KDE that doesn't allow them to be merged into one project?
1) Politics. GNOME is Free Software and KDE is Open Source. Personally, I can't tell the difference between the two, but independent experimentation has shown that if you put a hardcore FSF/GNU person in the same room as an Open Source advocate, you can actually warp the curvature of time/space. And besides, GNOME is a GNU project and KDE is not, and for some people that makes all the difference in the world.
2) Language. GNOME is C and KDE is C++. Both have bindings to other languages, but for all practical purposes, one is C and the other C++.
3) Toolkits. First off, you have Qt versus GTK+. Until you can merge these two into one toolkit, you will never be able to merge KDE and GNOME. Both have different APIs, different OO paradigms, and different scopes.
IMHO that's all that is required to finally get a solid Desktop presence for linux.
If it is required for Linux to have a monolithic core of elites deciding what software must be used, before it can get on the desktop, then perhaps the desktop is not the right goal.
But no matter to me, since I use FreeBSD instead of Linux. There's choice for you!
A quick check of WIDI shows that Debian gets a 26%. That's a far cry from the 48% this survey shows. As for the OSE study, Debian was one of their sample pools to start with, so of course it's going to be skewed in that direction.
Here's my minority report from off the deep end. It sounds like to me that you will have someone maintain the boxes, and that all the user will ever see is the desktop. Fine. Then try FreeBSD.
FreeBSD is very easy to administer and has all the software Linux has. Stability and security is your prime concern in a public environment like this, and FreeBSD holds its own here. Only a few Linux distros can compare in this area (and the for-the-masses distros aren't them).
Both Bush and Clinton won with a plurality, not a majority. Please pull out your dictionary.
Every sampling method is skewed, even door-to-door canvassing. But Geez Louise! Trying to draw meaningful conclusions from a 100% self-selected sample is the epitome of anti-statistics.
Next you're going to argue for the robustness of the Slashdot Poll, aren't you?
What is the statistical uncertainty on a self-selected sample of 2200? My guess is in the neighborhood of 2150...
This survey consisted entirely of self-selected participants. The surveyors actually seemed to boast that they didn't select anyone themselves. This study is bogus and proves nothing.
The staggering support for Debian (48%!) only proves that Debian developers are more successful in recruiting other Debian developers to participate in bogus online survey's. This isn't a dig against Debian, it's a dig against the silly methodology of this study.
This survey, despite its seemingly thoroughness, is no more valid than the weekly Slashdot Poll. The only conclusion that can be drawn from this survey is that Debian, GNOME and vi users are more likely to participate in a self-selected survey than Redhat, SuSE, Mandrake, KDE and emacs users.
Somehow I have a hard time believing that a bunch of developers who say "we won't follow any sensible UI design until we get mandatory guidelines" are the sort of people that would follow the guidelines to begin with.
What GNOME really needs is a release manager with the cojones to kick out anything that doesn't follow the standards.
I'm trying to remember Scott McNealy's exact quote at LWCE. Something like "Buy your Linux PC at Walmart instead of Dell. You'll get the same service."
But if the hardcore "compile or die" guys impose their view of computing on it, open source will fail fairly quickly
Who said anything about imposing our choice on you? We don't feel the need for a single unified free unix-like operating system. We don't think that only one distribution should come out on top. When we hear Linus say "world domination" we recognize that it is humor and not a goal.
I am perfectly happy with you using Mandrake just as long as you let me use Gentoo and FreeBSD.
Because it is.
I never used Mandrake long enough to get to this level of frustration. I left Mandrake a long time before that could happen. I suggest you do the same.
But I do agree somewhat with your basic premise. Mandrake seems to operate under the premise that "easy to use" means "no intermediate or advanced users allowed". Oh, advanced Mandrake users are fine, but if you have any knowledge of Unix, Posix or other distros, you're left outside in the cold.
The best thing about Corel LinuxOS was that it was small. Really. It asked you about three or four questions, then installed a very minimal default. You got a KDE desktop and that was it. On that desktop was an icon that allowed you to download a million other packages, but you didn't do that during the install.
I watched a greenhorn newbie install SuSE once. It was painful. He didn't have enough room for a complete full install, so he spent hours choosing between jed, joe, jove, emacs, xemacs, nvi, vim, elvis, pico and and a million other text editors. A newbie shouldn't have to choose between 7000 packages during the install. That is ridiculous.
An easy to use Linux installer will install just the basics necessary to get to a desktop. Even that minimal set of software is still going to give the user ten times the functionality of the full Windows install.
No, no, no! The United States is the Great Satan. We must attack it always with complete disregard to the truth. Automobiles are only driven in the US, polluted rivers only occur in the US, and only US oil tankers leak. This is the result of capitalism, which only exists in the US.
Back in my day if you went to college you had to take humanities and science classes. Stuff like foreign languages, composition, biology and philosophy were requirements. Even the engineering students. To make it fair, the liberal arts students had to take calculus as well.
But you tell that to youngsters nowadays and they won't believe you!
We give our code away to everyone, rich or poor, short or tall, stupid or smart. We don't attach strings to it. If you don't like Bill's copy of the code, download your own.
As a longtime BSD advocate, I must partly disagree. SELinux must be under the GPL, because it's based on GPLd software. The NSA can't arbitrarily change the license. As long as the choose to create derivative products from GPLd code, they must use the GPL. Besides which, public domain is incompatible with the GPL.
p.s. On the other hand, the NSA getting involved in a hardened BSD OS would be awesome.
but mostly talks about the much more benign stuff lumped under "direct marketing," like reminder updates from stores you cleared to send it to you.
"Hello, you are receiving this message because you selected to receive such messages on our website, one of our competitor's websites, or a completely unrelated website. If you do not wish to receive further messages of this type, please verify the validity of your email address by visiting the following address with a cookie-enabled browser. By removing your address from our list, you indicate your wish to receive similar messages of this type.
Because licenses are private agreements between private individuals. Both parties enter into this agreement voluntarily of their own free will.
Some may argue that certain proprietary licenses are not voluntary, but neither the BSD or GPL fall into this category.
Well maybe, just maybe, KDE didn't know about this bug until it was reported. Ever think of that?
Every change like that can have hundreds of ramifications
Well duh! That's why they didn't make a release ninety minutes later!
If the fix had remained on the developers desk, no one could have possibly tested it. But by committing it to CVS, the testing can begin.
As a professional software developer, I expect I could have done a moderate amount of unit testing before I committed a fix like this. But there's no way in hell anyone is going to be able to do any integration or systems test until I commit it to the code base.
Code review and QA testing occur in the CVS tree. Neither can happen until the code is committed. Or are you recommending that KDE start using a private, restricted and closed development environment?