When we're dealing with a monopoly, the rules are different.
Did you ever stop to think about that statement. Really think about it? Ever wonder how fscked the world would be if that attitude were applied universally?
Just because it is the law doesn't make it a good law. Removing the blindfold from Lady Justice is far too grave a matter to justify a separate standard for monopolies.
striking new evidence of what in my opinion can only be described as abuse of their monopoly position
Every piece of proprietary software that makes a profit is in exactly the same position. So why pick on just Microsoft? The monopoly isn't limited to Microsoft, but available to anyone that has a copyright.
I know it's fun picking on Microsoft, especially after some of the stupid stunts they've been pulling, but before you start waving evidence around, stop to think about what it means.
One example: Ximian Connector. Proprietary software. Currently selling for $69 single user. If Ximian can sell this piece of non-free software for $39 instead of $69 and still make a profit, then Ximian is a monopoly!
I've never used XP, or hung around with the XP crowd, so maybe the situation has changed in the last year. But last I checked the Windows GUI was an inconsistant hodgepodge merely pretending to be consistant. When every multimedia application under Windows has a radically different look-and-feel from each other, the claim of consistancy is silly.
The only "rulebook" for Windows interfaces is the choice of toolkit. If you use VS you use the VS rulebook. But if you use aonther tookit (borland), then you use a different rulebook. Or you can program in straight GDI/win32 and write your own rulebook.
People think Windows has a consistent interface because Microsoft keeps telling them that it is. But is certainly isn't. If every application you use is from Microsoft, it's better. But even then you have inconsistancies depending on the age of the application. Every release of MSOffice introduces new inconsistant UI elements.
To address one of your points in detail, lets look at icons. The toolkits are not in charge of the icons. They can be whatever the developer chooses to put in their resource file. The only reason so many of them seem the same is because some standard ones were included in the SDK and with VS. Before OWL got buried by MFC, it was a simple matter to tell VC++ apps from BC++ apps: the icons were subtly different. Today you can often tell Microsoft apps from non-Microsoft apps because MS uses their own icons that look a lot better than those they distribute to developers.
To come to your point, no, picoGUI cannont embed the X protocol (it would be against its basic approach). But il could be possible (though not easy) to make 'compatible library' that traslates GTK+ API (or QT API) into the picoGUI
Another solution, perhaps better, is to make an X toolkit that uses the picoGUI API. Then an application won't care if the system has picoGUI, or is administered by an old fart with X.
The entrenched X API is the biggest roadblock to projects like these. It's almost easier to put picoGUI into a new OS rather than trying to make it replace X in Linux/BSD/UNIX.
But if the X API can be put on picoGUI, then it's something to think about. Otherwise I'll stick with XFree86, which is getting faster and smaller with each release.
I really wish the nix community would quit trying to be leet and make an OS for the rest of us already.
Get a clue. We're not making an OS for you, we're making one for us! Get it? If you don't like it you can either become one of us, or make your own. Geez...
(oh, and in case anyone thinks im a newb, im currently in school for computers, have an A+, will soon have Net+ and w2k MCP, and i've been running linux for 6 months.)
Sounds like a newb to me... If you've got the skills, use them to improve freenix. Otherwise stop bitching about what we've sweated decades to make, and which we give to you for free.
Re:I can beat all the insect/reptile takeover stor
on
Ants Invade iBook
·
· Score: 1
I used to work in a floorcovering store. We pulled out a roll of carpet during the winter and discovered dozens of LIZARDS within the wraps. Eww! They crawled in there for the winter and were alive but dormant in the cold.
For me the problem has been that KDE and GNOME both crash on me and dump me out to a command prompt in a console.
I think your problem is your distribution. Just a guess. I have never had either KDE or GNOME crash on me. I think one of the up and coming problems with Linux is a few shoddy distributions giving the good ones a bad name.
1) Less hassle dealing with the OS. I don't care anything about the "OS" part when I'm using a machine. I use applications. Windows is far easier to install and use applications on than Linux. application and install break windows far less than on Linux IMHE.
True and false.
True, most people don't want to deal with anything lower level than the desktop and the applications that run on it. They shouldn't have to. But unfortunately there is reality to deal with. Reality says that if you are the administrator of a system, then you must to administrator type things. The only two solutions to this as I can see are a miraculous set of default settings that work for everyone, or user knowledge of how computers and networks work.
What Linux, BSD, UNIX needs to do is to come up with different distributions for different users. Some tentative steps have been taken in this direction, but no distro has gone far enough yet. If you give the user the power to run a secure server, you've given them enough rope to hang themselves.
False, on decent Linux or BSD system, installation of software is more robust and easier than under Windows. Packages are easier to use than InstallShield type programs. Packages are a better paradigm than embedded installers.
The problem comes when distros/packagers screw up their packaging. Do people blaim Windows when some random install.exe file craps out and hoses their AOL installation? No! So why do people blaim Redhat/SuSE/Mandrake/etc when some random RPM hoses another package? A few popular distros need to get on the ball and start making more robust packages, because all the good distros and systems are tired of sharing the blame.
2) The applications themselves.
Amen. Regardless of what people say about usability, common looks and feels, and handholding installers, the lack of professional quality applications in every field is what's holding freenix back. Mozilla and OpenOffice went a long ways, but we still need a lot more.
3) Development. Believe it or not developing for Windows is infinitely nicer than developing on Linux (Okay, that's just my opinion). The tools are all equal (gcc, perl, python, vi, emacs) up to far more advanced (Visual Studio) and far more varied to choose from.
I have to disagree. This is a matter of style, not advancement or variety. Tell me I have to develop under Windows and the first thing I will do is install XEmacs. If I don't have to use MFC, I won't even touch Visual Studio. It all depends on what you're used to and how you prefer to work.
So far I've read about a third of the posts. I can't read them all because there's a lot. But so far I have not found what I was expecting to see.
No one is claiming that they're staying on Windows because KDE and GNOME look different! There's this sense of urgency in the Linux community that unless there's a unified vanilla desktop, no one is going to want to use Linux. It seems that this is not the case.
But maybe I've missed those posts. So let me ask: is there anyone out there who has genuinely stayed with Windows precisely because KDE and GNOME don't have the same look and feel? [I'm not asking if you want them to have to same look, only if you have honestly refused to use any form of UNIX because of it]
Ditch X and you'll lose as many people as you'll gain. DirectX is better than X11 for direct access to the hardware buffer. If you're a game player, that makes a difference. X11 is better than GDI/DirectX/w32 for systems on a network. For the corporate environment, that makes a difference. For everything else, there's no difference in performance.
Because of X11, I can run Adobe FrameMaker on my FreeBSD machine. The application is running on a Solaris box, but I can interact with it as if it were native, with no noticable latency. I'll refuse any alternate solution that cannot do the same.
p.s. I've also got Win2K on the same system. I find that XFree86 is actually faster on the same hardware. I don't know how much of that is due to XFree86 as opposed to FreeBSD, but I don't care, because I'm not giving up XFree86.
No one should ever be forced to change their given name. Period. Why can't people like you understand that real people are living real lives, and that they names are not properties to be bought and sold via lawsuits.
There is absolutely NO inconvenience if both Bill Wymans use their own names. Geez! Next thing you know you'll be arguing that there should be only one John Smith in the world.
That also screws over a lot of people who left Windows development precisely because of the massive reliability on C++.
The people who left Windows solely because of C++ can be counted on one hand. If you don't like the MFC, use straight Win32 with C instead. Duh! Or use Java, Perl, Python, Lisp, etc. I even hear you can use Miguel's favorite byte-code language!
People leave Windows for much more substantial reasons than Visual Studio including C++ along with C, C# and VBasic.
Not bad, but pretty lame that it can't even do a single functional language, much less Objective C.
There is an Objective-C binding for Qt/KDE, it's just not finished, so it's not in the list. These bindings don't appear out of thin air, they need developers who use those languages to make them. If you have a favorite obscure language that needs Qt/KDE bindings, then get your ass on the ball and write some!
Note: Bindings for popular scripting languages is a plus. Qt supports the three most popular ones. It also supports the two most popular byte compiled languages. And since it supports C, you have your common denominator.
Gtk has [gtk.org] Ada, C++, Perl, Python, Common Lisp, Eiffel, Erlang, Guile, Haskell, Java, JavaScript, Objective-C, Objective-Caml, Objective-Label, Pascal, PHP, Ruby, TCL, TOM, and XBase bindings.
This is taking it to the extreme. Some of those languages are obscure. Others are inappropriate lanaguages for GUI application development.
GTK+/GNOME makes the mistake of trying to please all of the developers all of the time. When a developer says "give me Ocaml or go to hell", sometimes the best solution is to say "sure, go to hell." If you'll notice, GNOME does not include any Ocaml code in its core packages. If any developers submit some, GNOME would be correct to reject them. You do not want to burden the user with twenty different runtime libraries.
p.s. My goal is not to slam Ocaml. I am using it as an example only.
The Qt people evidently didn't see compatibility with less-used languages as important, so they snubbed everyone else.
When I publish documentation in English, I am not snubbing speakers of German, French or Cantonese. I publish my documentation in English because that is what I know. But I do not restrict translations. If you want my stuff in German, you have my blessings to go forth and translate.
Likewise, if you want Ocaml bindings for Qt/KDE, you have my blessings, along with Trolltech's, to go forth and bind.
Really? When did they lose the QString and QVector class and other ugly warts?
They did not lose them. But they made them STL compatible. That means you can use the STL instead of the QTL if you choose. Trolltech can't get rid of them now because some platforms and compilers still don't support the STL or modern templates. And the three major STL implementations are subtly different from each other, which plays havoc with a crossplatform library.
And besides which, while QVector may now be redundant, QString is not! QString has twenty times the power and flexibility of STL string.
"And it's bigger because it does a heck of a lot more than widgets-only GTK+"
Which is a *bad* thing, violating the UNIX design principle of many small parts, each the best for its job.
The "small parts" philosophy of UNIX applies to applications, programs and utilities. It does not apply to libraries. Go check out libc for an example of another library that includes everything but the kitchen sink.
Qt is a crossplatform application framework. It was written with a completely different set of goals than GTK+, which aimed to be just a widget toolkit.
And of course, it's a dynamic library, so your application only loads what it needs.
Qt is not a free beer library. And that's a sorry state of affairs.
No commercial software development is going to be profitable with free-beer as a product. The only choices in FOSS for profitable development libraries is to offer a hard-copyleft product, or offer a proprietary version for proprietary developers. Neither will be free-beer for proprietary developers. Maybe you should get off the FSF bandwagon and join the Free Beer Foundation instead.
The impression I got is that Qt pretty much exists to pander to ex-Windows types, who are used to coding in C++.
Where does this rumour come from that C++ == Windows? It's silly. C++ started out on UNIX. It was completed on UNIX. It was been used on UNIX since day one. Microsoft pushes Visual Basic more than Visual C++. It's now emphasizing C#. So why do people persist in equating C++ with Windows?
They only people Trolltech panders to are crossplatform application developers. Go read their 3.1 announcement. Half the stuff panders to crossplatform interests.
Trolltech chose C++ for Qt because a popular object oriented language is preferable for GUI and application development. C++ is by far the most popular object oriented language. It compiles into fast and efficient native code. They made the right choice.
I've heard that before, but I still find it silly. There are several levels to the POSIX standards, and NT/XP meet only a few of them.
The point i was making is that QT claims to be a cross platform dev environment of which KDE is the flagship so the natural step should be cross platform KDE.
Qt does more than claim, it IS an crossplatform development environment. It's the best I have every seen. But KDE requires more than Qt. It also requires X11, since some display stuff appropriate for a desktop is not appropriate for a widget library. It also assumes a POSIX and UNIX infrastructure.
Some stuff could be ported relatively easily though. Other stuff would be impossible or would need to be rewritten from scratch. Most stuff would be inbetween, and difficult but not impossible. I can envision a Konqueror port. I cannot envision a KWin or Kicker port.
I'm an Enlightenment and plan9 user anyway. I like my desktops empty not crammed with silly little pictures.
Last time I used Enlightenment, it had more silly little pictures than KDE ever dreamed of. Of course you can make Enlightenment as bare as plan9, but you can also make KDE equally spartan.
Qt supports a lot of languages. Sure, C++ is it's native and preferred language, but GTK+ has a native and preferred language to, known as C. Qt uses the STL. Complain to your package builders, because Trolltech didn't make them. Duh! And it's bigger because it does a heck of a lot more than widgets-only GTK+.
You don't think their widgets are attractive? Even when it has a native GTK+ widget theme (motif-plus)? Then go grab some others! Liquid, Keramik, Qinx, etc. Or write your own.
Some people have said that this needs to be the case for Qt to be funded -- well, gtk manages without putting annoying licensing into their product
GTK+ is a business? It earns revenue? Wow! When did that happen? GTK+ manages with the LGPL because it is a free beer library.
Qt versus VS: No comparison! They're completely different creatures. One is a crossplatform library, the other is Windows only. One is quality, the other crap. One is a full application framework, the other is a compiler with a few OS headers and wizards. One has a clean OO architecture, the other is chock full of unreadable macros.
What about shareware? If one shareware developer can't generate enough revenue to cover $1500, then it's time to look for a better business model. Geez!
Why not LGPL? Because Qt is of such high quality that no one needs support. If you make the product free-beer, where is the revenue going to come from? Trolltech isn't a charity. It's a business. It needs to generate revenue to stay in business. The only FOSS libraries making revenue are those that require support.
"write unit tests first." "leave optimization 'till last." "develop iteratively."
Yup. Doesn't sound like much, but these simple gems work. I don't know XP, and avoid anything with such trite nomenclature. But here is another tip good for any methodology:
So what? Play this trick question on a person, and when they answer wrong, and you give them the correct answer, observe their response. Invariably they will say "duh! I knew that!"
All this question will measure is a person's ability to filter out extraneous information. This may be a useful skill, but it is not a measure of intelligence of linguistic aptitude. Most certainly it does not measure an aptitude for Unix systems administration.
-----
Two hundred people die as an airplane full of French diplomats crashes exactly on the border between Germany and Belgium. Where do they bury the survivors?
Slashdot Mirror
When we're dealing with a monopoly, the rules are different.
Did you ever stop to think about that statement. Really think about it? Ever wonder how fscked the world would be if that attitude were applied universally?
Just because it is the law doesn't make it a good law. Removing the blindfold from Lady Justice is far too grave a matter to justify a separate standard for monopolies.
striking new evidence of what in my opinion can only be described as abuse of their monopoly position
Every piece of proprietary software that makes a profit is in exactly the same position. So why pick on just Microsoft? The monopoly isn't limited to Microsoft, but available to anyone that has a copyright.
I know it's fun picking on Microsoft, especially after some of the stupid stunts they've been pulling, but before you start waving evidence around, stop to think about what it means.
One example: Ximian Connector. Proprietary software. Currently selling for $69 single user. If Ximian can sell this piece of non-free software for $39 instead of $69 and still make a profit, then Ximian is a monopoly!
Tell me why it's ok for women to call us pigs but we can't make the occasional crack?
It's called "political correctness".
I've never used XP, or hung around with the XP crowd, so maybe the situation has changed in the last year. But last I checked the Windows GUI was an inconsistant hodgepodge merely pretending to be consistant. When every multimedia application under Windows has a radically different look-and-feel from each other, the claim of consistancy is silly.
The only "rulebook" for Windows interfaces is the choice of toolkit. If you use VS you use the VS rulebook. But if you use aonther tookit (borland), then you use a different rulebook. Or you can program in straight GDI/win32 and write your own rulebook.
People think Windows has a consistent interface because Microsoft keeps telling them that it is. But is certainly isn't. If every application you use is from Microsoft, it's better. But even then you have inconsistancies depending on the age of the application. Every release of MSOffice introduces new inconsistant UI elements.
To address one of your points in detail, lets look at icons. The toolkits are not in charge of the icons. They can be whatever the developer chooses to put in their resource file. The only reason so many of them seem the same is because some standard ones were included in the SDK and with VS. Before OWL got buried by MFC, it was a simple matter to tell VC++ apps from BC++ apps: the icons were subtly different. Today you can often tell Microsoft apps from non-Microsoft apps because MS uses their own icons that look a lot better than those they distribute to developers.
To come to your point, no, picoGUI cannont embed the X protocol (it would be against its basic approach). But il could be possible (though not easy) to make 'compatible library' that traslates GTK+ API (or QT API) into the picoGUI
Another solution, perhaps better, is to make an X toolkit that uses the picoGUI API. Then an application won't care if the system has picoGUI, or is administered by an old fart with X.
The entrenched X API is the biggest roadblock to projects like these. It's almost easier to put picoGUI into a new OS rather than trying to make it replace X in Linux/BSD/UNIX.
But if the X API can be put on picoGUI, then it's something to think about. Otherwise I'll stick with XFree86, which is getting faster and smaller with each release.
I really wish the nix community would quit trying to be leet and make an OS for the rest of us already.
Get a clue. We're not making an OS for you, we're making one for us! Get it? If you don't like it you can either become one of us, or make your own. Geez...
(oh, and in case anyone thinks im a newb, im currently in school for computers, have an A+, will soon have Net+ and w2k MCP, and i've been running linux for 6 months.)
Sounds like a newb to me... If you've got the skills, use them to improve freenix. Otherwise stop bitching about what we've sweated decades to make, and which we give to you for free.
I used to work in a floorcovering store. We pulled out a roll of carpet during the winter and discovered dozens of LIZARDS within the wraps. Eww! They crawled in there for the winter and were alive but dormant in the cold.
Bad idea. Very bad idea. That's not what XFree86 is for. Plus, making it dependent upon GNOME is seriously jacked.
For me the problem has been that KDE and GNOME both crash on me and dump me out to a command prompt in a console.
I think your problem is your distribution. Just a guess. I have never had either KDE or GNOME crash on me. I think one of the up and coming problems with Linux is a few shoddy distributions giving the good ones a bad name.
1) Less hassle dealing with the OS. I don't care anything about the "OS" part when I'm using a machine. I use applications. Windows is far easier to install and use applications on than Linux. application and install break windows far less than on Linux IMHE.
True and false.
True, most people don't want to deal with anything lower level than the desktop and the applications that run on it. They shouldn't have to. But unfortunately there is reality to deal with. Reality says that if you are the administrator of a system, then you must to administrator type things. The only two solutions to this as I can see are a miraculous set of default settings that work for everyone, or user knowledge of how computers and networks work.
What Linux, BSD, UNIX needs to do is to come up with different distributions for different users. Some tentative steps have been taken in this direction, but no distro has gone far enough yet. If you give the user the power to run a secure server, you've given them enough rope to hang themselves.
False, on decent Linux or BSD system, installation of software is more robust and easier than under Windows. Packages are easier to use than InstallShield type programs. Packages are a better paradigm than embedded installers.
The problem comes when distros/packagers screw up their packaging. Do people blaim Windows when some random install.exe file craps out and hoses their AOL installation? No! So why do people blaim Redhat/SuSE/Mandrake/etc when some random RPM hoses another package? A few popular distros need to get on the ball and start making more robust packages, because all the good distros and systems are tired of sharing the blame.
2) The applications themselves.
Amen. Regardless of what people say about usability, common looks and feels, and handholding installers, the lack of professional quality applications in every field is what's holding freenix back. Mozilla and OpenOffice went a long ways, but we still need a lot more.
3) Development. Believe it or not developing for Windows is infinitely nicer than developing on Linux (Okay, that's just my opinion). The tools are all equal (gcc, perl, python, vi, emacs) up to far more advanced (Visual Studio) and far more varied to choose from.
I have to disagree. This is a matter of style, not advancement or variety. Tell me I have to develop under Windows and the first thing I will do is install XEmacs. If I don't have to use MFC, I won't even touch Visual Studio. It all depends on what you're used to and how you prefer to work.
So far I've read about a third of the posts. I can't read them all because there's a lot. But so far I have not found what I was expecting to see.
No one is claiming that they're staying on Windows because KDE and GNOME look different! There's this sense of urgency in the Linux community that unless there's a unified vanilla desktop, no one is going to want to use Linux. It seems that this is not the case.
But maybe I've missed those posts. So let me ask: is there anyone out there who has genuinely stayed with Windows precisely because KDE and GNOME don't have the same look and feel? [I'm not asking if you want them to have to same look, only if you have honestly refused to use any form of UNIX because of it]
Ditch X and you'll lose as many people as you'll gain. DirectX is better than X11 for direct access to the hardware buffer. If you're a game player, that makes a difference. X11 is better than GDI/DirectX/w32 for systems on a network. For the corporate environment, that makes a difference. For everything else, there's no difference in performance.
Because of X11, I can run Adobe FrameMaker on my FreeBSD machine. The application is running on a Solaris box, but I can interact with it as if it were native, with no noticable latency. I'll refuse any alternate solution that cannot do the same.
p.s. I've also got Win2K on the same system. I find that XFree86 is actually faster on the same hardware. I don't know how much of that is due to XFree86 as opposed to FreeBSD, but I don't care, because I'm not giving up XFree86.
That was so funny I think I just made a skid mark!
No one should ever be forced to change their given name. Period. Why can't people like you understand that real people are living real lives, and that they names are not properties to be bought and sold via lawsuits.
There is absolutely NO inconvenience if both Bill Wymans use their own names. Geez! Next thing you know you'll be arguing that there should be only one John Smith in the world.
Wrong! Bill Wyman the author should not have to change his name, give up his domain, or pay any sort of monetary homage to Bill Wyman the bassist.
Assigning any sort of property rights to a person's *name* is assinine. Even as a trademark.
That also screws over a lot of people who left Windows development precisely because of the massive reliability on C++.
The people who left Windows solely because of C++ can be counted on one hand. If you don't like the MFC, use straight Win32 with C instead. Duh! Or use Java, Perl, Python, Lisp, etc. I even hear you can use Miguel's favorite byte-code language!
People leave Windows for much more substantial reasons than Visual Studio including C++ along with C, C# and VBasic.
Not bad, but pretty lame that it can't even do a single functional language, much less Objective C.
There is an Objective-C binding for Qt/KDE, it's just not finished, so it's not in the list. These bindings don't appear out of thin air, they need developers who use those languages to make them. If you have a favorite obscure language that needs Qt/KDE bindings, then get your ass on the ball and write some!
Note: Bindings for popular scripting languages is a plus. Qt supports the three most popular ones. It also supports the two most popular byte compiled languages. And since it supports C, you have your common denominator.
Gtk has [gtk.org] Ada, C++, Perl, Python, Common Lisp, Eiffel, Erlang, Guile, Haskell, Java, JavaScript, Objective-C, Objective-Caml, Objective-Label, Pascal, PHP, Ruby, TCL, TOM, and XBase bindings.
This is taking it to the extreme. Some of those languages are obscure. Others are inappropriate lanaguages for GUI application development.
GTK+/GNOME makes the mistake of trying to please all of the developers all of the time. When a developer says "give me Ocaml or go to hell", sometimes the best solution is to say "sure, go to hell." If you'll notice, GNOME does not include any Ocaml code in its core packages. If any developers submit some, GNOME would be correct to reject them. You do not want to burden the user with twenty different runtime libraries.
p.s. My goal is not to slam Ocaml. I am using it as an example only.
The Qt people evidently didn't see compatibility with less-used languages as important, so they snubbed everyone else.
When I publish documentation in English, I am not snubbing speakers of German, French or Cantonese. I publish my documentation in English because that is what I know. But I do not restrict translations. If you want my stuff in German, you have my blessings to go forth and translate.
Likewise, if you want Ocaml bindings for Qt/KDE, you have my blessings, along with Trolltech's, to go forth and bind.
Really? When did they lose the QString and QVector class and other ugly warts?
They did not lose them. But they made them STL compatible. That means you can use the STL instead of the QTL if you choose. Trolltech can't get rid of them now because some platforms and compilers still don't support the STL or modern templates. And the three major STL implementations are subtly different from each other, which plays havoc with a crossplatform library.
And besides which, while QVector may now be redundant, QString is not! QString has twenty times the power and flexibility of STL string.
"And it's bigger because it does a heck of a lot more than widgets-only GTK+"
Which is a *bad* thing, violating the UNIX design principle of many small parts, each the best for its job.
The "small parts" philosophy of UNIX applies to applications, programs and utilities. It does not apply to libraries. Go check out libc for an example of another library that includes everything but the kitchen sink.
Qt is a crossplatform application framework. It was written with a completely different set of goals than GTK+, which aimed to be just a widget toolkit.
And of course, it's a dynamic library, so your application only loads what it needs.
Qt is not a free beer library. And that's a sorry state of affairs.
No commercial software development is going to be profitable with free-beer as a product. The only choices in FOSS for profitable development libraries is to offer a hard-copyleft product, or offer a proprietary version for proprietary developers. Neither will be free-beer for proprietary developers. Maybe you should get off the FSF bandwagon and join the Free Beer Foundation instead.
The impression I got is that Qt pretty much exists to pander to ex-Windows types, who are used to coding in C++.
Where does this rumour come from that C++ == Windows? It's silly. C++ started out on UNIX. It was completed on UNIX. It was been used on UNIX since day one. Microsoft pushes Visual Basic more than Visual C++. It's now emphasizing C#. So why do people persist in equating C++ with Windows?
They only people Trolltech panders to are crossplatform application developers. Go read their 3.1 announcement. Half the stuff panders to crossplatform interests.
Trolltech chose C++ for Qt because a popular object oriented language is preferable for GUI and application development. C++ is by far the most popular object oriented language. It compiles into fast and efficient native code. They made the right choice.
It's already posix compliant.
I've heard that before, but I still find it silly. There are several levels to the POSIX standards, and NT/XP meet only a few of them.
The point i was making is that QT claims to be a cross platform dev environment of which KDE is the flagship so the natural step should be cross platform KDE.
Qt does more than claim, it IS an crossplatform development environment. It's the best I have every seen. But KDE requires more than Qt. It also requires X11, since some display stuff appropriate for a desktop is not appropriate for a widget library. It also assumes a POSIX and UNIX infrastructure.
Some stuff could be ported relatively easily though. Other stuff would be impossible or would need to be rewritten from scratch. Most stuff would be inbetween, and difficult but not impossible. I can envision a Konqueror port. I cannot envision a KWin or Kicker port.
I'm an Enlightenment and plan9 user anyway. I like my desktops empty not crammed with silly little pictures.
Last time I used Enlightenment, it had more silly little pictures than KDE ever dreamed of. Of course you can make Enlightenment as bare as plan9, but you can also make KDE equally spartan.
Hah! Get a clue dude!
Qt supports a lot of languages. Sure, C++ is it's native and preferred language, but GTK+ has a native and preferred language to, known as C. Qt uses the STL. Complain to your package builders, because Trolltech didn't make them. Duh! And it's bigger because it does a heck of a lot more than widgets-only GTK+.
You don't think their widgets are attractive? Even when it has a native GTK+ widget theme (motif-plus)? Then go grab some others! Liquid, Keramik, Qinx, etc. Or write your own.
Some people have said that this needs to be the case for Qt to be funded -- well, gtk manages without putting annoying licensing into their product
GTK+ is a business? It earns revenue? Wow! When did that happen? GTK+ manages with the LGPL because it is a free beer library.
When can I expect a [Windows] native KDE?
When Windows becomes a UNIX. Until then the effort spent porting to an archaic architecture isn't worth it.
Qt versus VS: No comparison! They're completely different creatures. One is a crossplatform library, the other is Windows only. One is quality, the other crap. One is a full application framework, the other is a compiler with a few OS headers and wizards. One has a clean OO architecture, the other is chock full of unreadable macros.
What about shareware? If one shareware developer can't generate enough revenue to cover $1500, then it's time to look for a better business model. Geez!
Why not LGPL? Because Qt is of such high quality that no one needs support. If you make the product free-beer, where is the revenue going to come from? Trolltech isn't a charity. It's a business. It needs to generate revenue to stay in business. The only FOSS libraries making revenue are those that require support.
Yup. Doesn't sound like much, but these simple gems work. I don't know XP, and avoid anything with such trite nomenclature. But here is another tip good for any methodology:
"Write the documentation first"
Agreed, libraries should be safe environment for children where they can walk past the computers without seeing porn.
Slashdot should be a safe environment for geeks where they can read posts without seeing the goatse man.
So what? Play this trick question on a person, and when they answer wrong, and you give them the correct answer, observe their response. Invariably they will say "duh! I knew that!"
All this question will measure is a person's ability to filter out extraneous information. This may be a useful skill, but it is not a measure of intelligence of linguistic aptitude. Most certainly it does not measure an aptitude for Unix systems administration.
-----
Two hundred people die as an airplane full of French diplomats crashes exactly on the border between Germany and Belgium. Where do they bury the survivors?