Trolltech has released Qt 4.0 both under commercial and GPL licenses for X11, Mac OS X and MS Windows. It is the first time that a MS Windows GPL edition is available. To celebrate the release Trolltech employees have created a song and a music video (Bittorrent download, Ogg Theora version). Read the Qt 4 Overview and the online Qt Reference Documentation for more information. You can download Qt from ftp.trolltech.com or from one of its mirrors. Work on KDE 4 has already started with making a development branch of KDE compile and run with Qt 4.
Unlike previous Qt releases, Qt 4 is a collection of smaller libraries which also allowed the restructuring of commercial offers into Qt Console for non-GUI development, Qt Desktop Light replacing the Professional and Qt Desktop comparable to Enterprise editions.
There are five new technologies that are new within Qt 4:
- Tulip, a new set of template container classes.
- Interview, a model/view architecture for item views.
- Arthur, the Qt 4 painting framework.
- Scribe, the Unicode text renderer with a public API for performing low-level text layout.
- Mainwindow, a modern action-based mainwindow, toolbar, menu, and docking architecture.
In addition, the following modules have been significantly improved since Qt 3:
- A fully cross-platform accessibility module, with support for the emerging SP-API Unix standard in addition to Microsoft and Mac Accessibility.
- The SQL module, which is now based on the Interview model/view framework.
- The network module, with better support for UDP and synchronous sockets.
- The style API, which is now decoupled from the widgets, meaning that you can draw any user interface element on any device (widget, pixmap, etc.).
- Enhanced thread support, with signal-slot connections across threads and per-thread event loops.
- A new resource system for embedding images and other resource files into the application executable.
The Qt Designer user interface design tool has been rewritten as a collection of interchangeable components. It now features support for MDI and SDI modes and supports custom widgets. The project editor and the code editor have been dropped.
Qt 3 based applications have to be ported to be able to run with Qt 4. A porting guide, a porting tool as well as a Qt3Support library for obsolete classes are provided. Trolltech aims to maintain the Qt3Support Library for the lifetime of the Qt 4 series, and will also support the Qt 3 series for a minimum of two years beyond the release of Qt 4.
Some known issues are listed which are expected to be fixed in upcoming maintenance releases of Qt 4.0. Qt 4.1 will features certain advanced Qt 3 features rewritten for Qt 4 which are now only available in the Qt 3 support library (eg. Qt 3 canvas, Qt 3 syntax highlighter). It is planned be released late in 2005.
Comments
>Personally, I only use windows at work 'cause I have no option.
>
I feel you all too well. HOWEVER, what is it you're lacking? OOo has opened and write to (via Linux) any Company's files I have needed too (and export them to PDF if need be - for free). File sharing? Samba. Email? If it's just Exchange you have to work with, there's plugins for that. If you one of the lucky that are straight imap or even pop, then I don't see a problem there. Is there something I'm missing?
M.
P.S.
This is *NOT* a flame. I'm HONESTLY very curious what it is from holding you back.
Not to be rude, but I suspect it's the lead system administrator.
Not all of us can do what we want you know!
More seriously, there are more serious software issues than the ones you're thinking of. A lot of (surprisingly large) businesses use things like Access & Visual Basic or Lotus Notes to manage their data - there's no easy way of carrying that across to Linux, and the price of bespoke software is always going to be high. Also, some business rely on Windows only tools. Further, they often use fairly complex accounting software, for which equivalents are not readily available for Linux (KMyMoney is not an option!), and tools like Crystal Reports for complex reporting.
Event companies doing bespoke development may need Windows tools as their clients will use Windows: this means Delphi, C++ Builder or VC. Kylix is not a viable option (it's unsupported and alledgedly buggy) and the cost of Qt/Win, Qt/X11 and a solid Linux distro is greater than Qt/Win and Windows. Doing desktop development with Java takes too long (although the upcoming Netbeans may rectify that), and Java on the desktop has a bad rep which may put off customers.
The only way you could switch easily to Linux is if you're doing Java development, and use open-source tools for change-control and issue-tracking like Subversion and Mantis. For most "normal" organisations, a switch is a major deal. It's never just Office and email.
Right on target. I work for one of the largest financial organizations globally. I can't possibly choose what software I run, and what OS I run. I work within a mid-sizegroup, and development is done in winXP boxes. It is horrible, multi-tasking is slow as sh*t (evne with a dual processor fast box). But these decisions are not made by me as an individual, but at a higher level, taking many factors into account. We do use Linux, and we are moving the client side of what we do to Java, and if at some point all our clients here use the Java interface, we could possibly move 100% to Linux, at least my group. We'll see, it looks very unlikely ...
The only OS stuff I use is where I get to choose (Firefox, Cygwin for some shell stuff, Kdiff3 which provides Win* binaries, and that's about it)
I think PyQt are going to become the real big winner because of this. It already looks like the PyQt use on X11 are increasing rapidly, adding Windows will probably multiply PyQt usage tenfold. A GPL, or proprietary if you so choose, RAD for all the major computing platforms(Win/X11/Mac) easily available, are going to get noticed. I'm looking forward to seeing the Python/PyQt/Designer/Eric3 combos starting to appear on CD/DVDs for mainstream windows magazines.
me and my girlfriend - we loved the musicvideo... :D
Perhaps your name gives a hint as to why :)
i have no idea what you are talking about :D
Weird. Mine loves it too. I think it's demeaning for computers, who are forced to reproduce it. But I kinda like it, too!
It is said so many times already, im gonna say it again: thank you trolltech for GPL'ing Qt/Win.
Free Qt bindings (PyQt, etc.), some (maybe all some day) KDE technologies, generally more GPL software for win32...
Im stuck with one more question: will more GPL software for win32 be good for out beloveth FreeSoftwareMovement? (are there developers to inspire on the win32 platform?)
_cies.
It will be good when it makes people start questioning the Microsoft fee they nearly always pay when buying a computer.
And it will be awesome for people like me that still have XP installed (FC3 only laptop, and dual boot Gentoo/XP on desktop, now most of the time use Gentoo on the desktop since my laptop makes my room to hot, I only reboot into Windows to play Warcraft 3 now).
In the past I had installed KDE-CygWin and used that for a while (pretty damn fast for being emulated), but if KDE apps are ported to native windows (like Konqueror and KIO-Slaves) then that will be REALLY good (I can install Konqueror on people's computer, and instead of using a web based file manager they can login via SFTP or FISH or worst case FTP).
It's really hard to grasp all the amazing technologies in KDE (KIO-Slaves alone is amazing), even after using KDE for several years now (since RedHat9, and for a few weeks with Mandrake 7.1).
Warcraft III is one of the first things I tried with Cedega. It works just as fast under Linux without the Window.
> will more GPL software for win32 be good for out beloveth FreeSoftwareMovement?
Yes, as Free Software is not about spreading Linux (or KDE).
No, as Free Software is not about promoting/supporting a proprietary, closed-source platform (Windows) that is hostile towards Free or Open Software.
If you would be right then there is something missing in Free Software licenses which prohibit Free Software to run on "proprietary, closed-source platforms". Free Software is about promoting Free Software and that is not always a OS kernel or the desktop but can be also about individual applications.
GNU has worked well on any kind of operating systems.
If GNU had been restricted to free open-source operating systems, that would have been hard and, probably, we would not be here discussing about it.
Have a nice day!
> If GNU had been restricted to free open-source operating systems, that would have been hard and, probably, we would not be here discussing about it.
In the beginning that was an evil that had to be accepted. But it's also the reason why Hurd was started.
oh look, my favourite topic ;-P but i do feel the need to make a clarification here:
it's not about open vs closed operating systems. please don't descend into that irrelevant land of straw men.
no, it's about environments that are good for Open Source / Free Software and ones that are a trap for it.
the only good thing that can come of Free Software on Windows for open source is if we manage to snag the minds and hearts of Windows developers and they start developing Free software. Free Software on Windows can do all sorts of wonderful things for the users of Windows, the question is how to turn that around to ensure Open Source development continues to thrive and most importantly grow.
if you think that the open source desktop can continue to exist in a meaningful way (e.g. AmigaOS or BeOS doesn't exist in a meaningful way anymore) forever with single percentage market share, then perhaps you have more faith in the good will of the desktop industry.
IMHO the most important applications we can port to windows are kdevelop, umbrello, quanta and kexi. tools people use to build tools.
btw, the whole "but open source apps on Windows lead to migrations" theory has been pretty well tested at this point with OOo and Firefox and shown to be mostly hot air. it's not commodity apps that keep people away from Open Source friendly platforms.
i understand the convenience factor for users who love KDE but are stuck on Windows at work or play, but convenience is not a useful metric in and of itself. think of other things in life that are convenient, like dropping garbage on the ground rather than in a trash can, but which aren't really all that great at a larger level.
It depends on how you see it.
To go back again to the GNU tools, there are alive because there are so many ports of them. (I have found that there are even some ports for Sinclair QL.)
But by restricting the supported operating systems (or enviroments if you do not want it restricted to the OS), you do not get a base for a further live of the software.
Sure I would prefer that there would be more computers running with open source environments but I know that it is barely a thing of technical superiority but that it is about public relation stuff.
Also I do not expect that active feedback will increase because mostly people wanting to give feedback could already finding ways to give feedback today.
As for ecology, sure it is important too, but as open source, it is not mainstream either.
Have a nice day!
let's try this one more time:
it's not about closed vs open platforms
it's about supportive vs hostile platforms
and it's not about PR, either. it's about sustainability.
now, can you please explain to me what about the above is so hard to understand that after having repeated it for how long now people such as yourself still insist on discussing it as if it were a matter of PR to do with open source versus closed source platforms? what can i do to help make it obvious enough that it sinks in and the conversation can be had on the right basis?
the title sais it all...
> it's about environments that are good for
> Open Source / Free Software and ones that
> are a trap for it.
The FreeSoftwareMovement depends on developers more that users (I seem to rememer you also saying that). So will the efford of port lead to new inspired developers, or new lame-complaining-users (that: install, complain, move away)
that's indeed the question...
in the end i think someone will port kde over to the smelly platform. we will see what it brings eventually anyway.
_cies.
I don't think the migration theory has been sufficiently tested yet, for two reasons:
1: If %OFFICE_SUITE%, %BROWSER%, and %EMAIL_CLIENT% are the only apps anybody uses, then OOo and the Mozilla family *should* be leading the charge. But the ecosystem is so much larger than that. And it's usually %OBSCURE_LITTLE_WIDGET_I_CANT_LIVE_WITHOUT% that adds a disproportionate amount of inertia. Most people can figure out MS Office->OOo, IE->Firefox, and maybe Outlook Express->Thunderbird, but not much else.
I think there's an opportunity there for a sort of Migration Tracker that scans your Windows environment, and evaluates what it would take to make The Big Switch. Maybe it can say "You can switch now! Here's the distros that include everything you need.", or "There's no equivalent for this yet, but if you switch to these open source Windows apps now, the rest of your system will be ready once we fill in the missing piece."
2: The other side of all that is another opportunity: A Linux distro that *makes the migration for the user*, instead of "Resize partitions to make room for / and /swap, and I hope your Windows partitions are FAT32!" Wouldn't it be nice if a fresh install of Fedora picked up your user name, network settings, bookmarks, My Documents folder, iTunes library, wallpaper, etc. from the Windows partition that it just shrank, and set them all up for you for your first boot into Linux?
Anybody in this audience is probably willing to do the under the hood work necessary to make Linux work for themselves. But we're still a long way from the "Mom puts the CD in the drive and clicks GO" stage.
Haven't gotten it to compile yet on windows. I extracted it and ran configure, which shows me the license and then asks if I accept. I type 'y' and enter and it just boots me back to the command line without doing anything or displaying anything.
Anyone got this to work yet? I have VS.NET 2003..
OpenSource version will not work with VS.NET 2003
You need to use mingw with OpenSource Qt
I have no idea how they will top that video for Qt 5.
So disturbing. I'd rather not think about it yet can't help thinking about it. :-)
Hope QT4 lay the foundation for a rock solid KDE4 based Linux desktop that can change and SAVE the world. Congratulations to Trolltech, may your business flourish!
I also want to thank trolltech. Whenever a company or fellow consultant is looking for a c++ development kit, I always point them your way in large part because I appreciate businesses that work hard to do the right thing.
I will try to continue to support you into the future. Thanks for all the wonderful technology.
KDE 4 will take a while, but I have no doubt that it will rock!
I'm quite curious. I really like KDE. Whenever I use linux, I use it as my desktop environment. Does the release of a win32 version of gpl qt mean I might soon be able to replace Explorer on my windows machine with KDE?
I don't know if someone will port Konqueror, for sure it will not be the first application to be ported. Kontact has higher priority and likely someone will write a small khtml-based Qt Web browser anyway soon.
I don't really care about the timeframe if it's going to happen -- it's been ten years that I've been staring at the same windows interface, another couple won't kill me. I'm just curious about whether it could or will actually happen.
>it's been ten years that I've been staring at the same windows interface, another couple won't kill me.
Damn boy! That one nearly killed me! 8-D
/me chuckles on
~Macavity
The long term damage of using Windows that long *will* be irrecoverable, and yes, it *will* lead to a horrible death. Another couple of years will probably shorten the years you have to suffer though. So save yourself the pain and be sure to use Windows every day and get greenhorn as it is released.
Hi, a question:
In old Qt versions the Designer managed prj. files, but now it seems that these functionality has been dropped?
Now I have to manage 'em myself via text editor?
Is there a reference on which keywords are available for .prj files?
QMake is a powerful tool I don't want to miss!
>8^) katakombi
Read the documentation of qmake?
Great! I like it!
On Windows Kdevelop plus GNU Compiler Collection would be a real Visual Studio Killer.
Indeed - I really miss KDevelop when working with GCC on Windows. The recent 3.2 version is very good. I have tried Dev-C++ and Eclipse on the Windows platform, but they don't even come close.
Same here, Quanta Plus and Kdevelop are really missed on windows plataform.
Don't take me wrong, I use Linux every day on my home computer, but at my university (that have like, but very old and badly configured, and I don't have the root pass) it would be really be good to spreat this kind of software, there are a lot of pcs already with firefox ;)
Beautiful. Absolutely beautiful. You guys are fantastic.
I don't think that trolltech have emphasized this enough: now with the GPL'd QT4 there will be an avalanche in the number of KDE/QT freesoftware available natively under windows. Not all, especially the ones that relay on Posix/Unix APIs which would continue to rely on Cygwin. IMO, this QT4 GPL would represent a pivot in the history of software for years to come. It is now, and only now, that we can effortlessly develop native c++ applications that can be compiled and run on all platforms; that marrigged with FOSS would be the killer technology for Desktop applications.
And has everyone already found the easter egg in Qt 4 Designer? :-)
Err... no. Care to explain this...?
What to explain? I will not tell you how to activate it :-)...
Found =)
And, i must say this.. you've a good photograph there, isn't it? Nice business cards ;)
But then again, I'm blind and deaf.
IMHO is would be great if KDE4 could be out in early 2006. IIRC KDE 3.5 is
planned to come in september, and pretty much of Qt4 porting will be done by that
time (as most devs run both branches), no?
So don't let wait your users toooo long (as it happened with KDE 3.2).
Btw I compiled today Qt4 on an AthlonXP2200, and it took 2 hours to compile, and
the demos run a little sluggish... :-(
I have an Nvidia GForce 5200 with the latest binary driver on Linux 2.6.12.
> IMHO is would be great if KDE4 could be out in early 2006.
Keep on dreaming...
> IIRC KDE 3.5 is planned to come in september
There is no schedule for KDE 3.5 yet.
> and pretty much of Qt4 porting will be done by that time
KDE 4 is about much more than just porting to Qt 4.
There was the idea to have KDE4 on KDE's 10th birthday in October 2006
As for KDE 3.5, it seems that it will be for this autumn. For September, it i too early, as August is a lost month for translations, as too many translators are taking their holiday.
Have a nice day!
I don't see anything about the PostScript problems:
Wrong font names in PS files.
TrueType fonts rendered as Type 42 bitmaps in PS files.
I don't see anything about the WYSISYG problem:
Now WYGIWYS on the screen but, the fonts on the screen are hinted at the
screen resolution and the fonts that are printed have the actual font
metrics. The result is too much space between glyphs -- sometimes the
spacing is uneven as well.
Do you have to troll everytime? Even if fixed, do you think they mention every bugfix as highlight of a new release?