Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
KDE GUI

KDE 1.1 is out 269

erich@wrq.com was the first to tell us that KDE 1.1 is released.Update: 02/07 10:38 by H : It appears that while the annoucement has been made the binaries are not yet availible-should be up any time now. Update: 02/07 02:55 by CT : thanks to Christian Kreibich for the KDE Logo.
This discussion has been archived. No new comments can be posted.

KDE 1.1 is out

Comments Filter:
  • by Anonymous Coward
    Is there Ada support yet?
  • by Anonymous Coward
    Very nice. Good work.
  • by Anonymous Coward
    KDE is a nice idea, but must Linux play "Windows wanna be"? I think not!!!!!
  • by Anonymous Coward
    Let's say I wanted to develop Qt. I'd want to work like Linus Torvalds develops Linux. That means:
    • People send me patches.
    • Every week I make a new patch against the previous version of my code.
    • Every week I make a new tarball containing plain source code for my new version. (no patches)
    • Nobody is discriminated against. The library is free for use with software that is GPL, BSD, LGPL, MPL, Artistic, and proprietary.
    AFAIK, only BSD and LGPL code is OK for this. Not even the GPL itself is good enough.
  • by Anonymous Coward
    The GPL should never be used for libraries.
    Libraries should be LGPL, although the BSD
    licence is OK too. Maybe the MPL is OK.
  • by Anonymous Coward
    I really enjoy using KDE 1.1pre2 from time to time, but occasionally I think it looks hampered by QT's widget set. Any developers have a review of QT2.0 beta that they'd like to post? It sounds like a great step, especially with that scrolling mouse wheel support.
  • by Anonymous Coward
    I am compiling pre2 right now, heh. Oh well :)

    I don't see any problems with the new QT license. And GNOME just is not ready yet. Once it is, I'll give it a serious try. Until then, I'll stick with KDE.

    Besides, Linux with KDE is 1000 times better than Microsh*t Winblows98...
  • by Anonymous Coward
    Is it threadsafe? If so, in what manner? Also, how does it handle printing?
  • by Anonymous Coward
    Man. This thing is beautiful.
  • by Anonymous Coward
    Opinions on the KDE mailing list are out.

    I have about 10 serious KDE 1.1pre2 bugs that I was about to report on the
    bug report site, that I don't see anyone else having reported, just
    lacking the time to put them in. Yes, releasing it now is premature. I
    sure hope there will be a 1.1.1 maintenance release soon.
  • by Anonymous Coward
    I think that when coolo says that there will be KDE 1.1.x he refers to the fact that the next development versions will be VERY experimental because of the move to QT 2.0 and especially CORBA/KOM. It is a major branch and it might take months to stabilize. Interesting times ahead.

    Nevertheless stability-wise 1.1 is much more stable than 1.0 and more polished. I don't think that there will be a need of a brown paper-bag version :-) However, more than 100 programs go into this release and some of them may be more complete than others. There will probably be problems on "other Unixes" where the tester base was small.

    Lotzi B.
  • by Anonymous Coward
    I've been working on QT2.0 styles for KDE1.2/QT2.0 and have come up with this so far:

    http://www.jorsm.com/~mosfet/qtnext-highcolor.gi f
    http://www.jorsm.com/~mosfet/qtnext-fileopen.gif

    mosfet@jorsm.com
  • by Anonymous Coward

    This is shaping up to be a very interesting arms
    race. Does anyone have the inside scoop on who
    started the move to theme technology? I first
    heard of GTK implementing themes, but did QT
    have this in the works for some time? (And
    please no mentions of: "Yeah my brother's tool
    kit did that bad in 1985.")

    Obviously QT had the base for different themes
    with both their Motif/Windows toggle but my
    guess is it was rather hard to implement and
    2.0 is a big re-working of that code.

    What both GTK and QT need is for data-binding
    to ODBC or other data sources? Businesses, the real
    pushers behind standards in these areas, need
    database applications. I think if Troll
    had put some people on database stuff instead
    of OpenGL widgets they would have businesses
    eating their product up.
  • Don't use 1.1pre2 !
    There are bugs in it (e.g. removing a symlink to a
    directory removes the directory itself, recursively !)
    And I fixed some other things in kfm.
    Please, please, when you can (binary packages are not
    available yet), upgrade to the real 1.1

    David Faure, faure@kde.org, kfm maintainer
  • ...my sides hurt from laughing...

    --

  • It's more likely that you're just some lamer who is talking out their ass. Get real. If you want people to take you serious, don't post AC. If you really have contributed to the kernel, Gimp, etc., like you claim, you wouldn't hide behind your ACness.

    I'm not opposed to ACs in general... people just shouldn't be expected to be taken very seriously if they post as AC.

    Oh, and I admittedly haven't made any significant contributions to and major OSS projects. I have however taken advantage of the freedom given to my by using OSS and have modifided serveral apps to better suit me and to fix bugs... one of them was even a KDE app.

  • The GPL would do good for what Troll Tech wants to do. Granted, important libraries such as GNU libc and the X libraries should be released under a more forgiving licence such as the LGPL, BSD-style, or X-style licence. But the GPL does good for libraries that you don't want proprietary developers to use without getting a special licence. Even Richard Stallman agrees with that (except for the "special licence" part of course).
  • A GPL'ed library could be used commercially as long as that commercial product was GPL'ed as well. That's good in some ways, but most companies wouldn't want to open up their code, and thus Troll would make considerably less money off it.

    The QPL allows them to keep food on their tables, and Open Source people to keep the source open.

  • If you're really anxious to get it:
    upload.kde.org, /pub/kde/Incoming

    Or wait a day or so for them to move it to the normal distribution directories.
  • It would be nice if Troll would GPL Qt, but they cannot. It is how they make money. They are being very generous by changing their license to make it conform to open source, and I thank them for it.
  • by Matrix ( 290 )
    Seeing how Qt looks like Motif and Lesstif looks like Motif ... Hm...
  • They're doing no harm at all. You're not forced to use it, are you? No, I didn't think so.
  • No kidding ;) don't ask me, I spend my time playing with a Window Maker desktop ;)
  • ...he's Linus!
    (just kidding :) )
  • It almost sounds reasonable that Linux would have been started in "Visual C" or "Masm" only if you're a complete uneducated fool- which is why so many of us responded along the lines of "LOL ROFL stop! stop! it hurts! ROFL!"
    It _is_ worth noting, however, that people without a clue might be conned by this sort of thing, and so it's good to note formally that the very suggestion is beyond ludicrous and in fact impossible. VC++ is not necessarily a more reliable tool- even if it were, which is debatable, it is absolutely certain that Linus used no such thing. Unix C compilers have been around for ages, since before X even... I daresay you could even look up what Linus had back then, it's not at all subject to doubt.
  • I don't use Gnome at all, as near as I can tell. If I wanted a desktop I'd boot into MacOS... but the times I have used KDE, my personal experience has not been of functionality, work and working software.
    My experience was of a thing that tried really hard to be just like Windows, had lots of little behaviors like what you'd get with Windows, but was horribly beset with minor glitches everywhere. Programs would launch a couple times and then be unlaunchable. Windows would freeze up- and KDE actually took pains to make sure I couldn't use PPP config information in nonKDE tools like wmppp, by literally making the important config info (in resolv.conf) _temporary_ and deleting it after use so no other program can use it unless it goes through kppp.
    I didn't buy a whole other disk and set up a working, live, genuine Linux installation to put up with crap like that.
    Again- I don't use Gnome and have had no real interest in trying to nail a desktop onto the side of X. But, I have used KDE, and in fact have hunted down some of the things it does, and my experience with it has been quite negative in every way. Most notably, I contest the implication that KDE is functionality, good design and working software. There is a lot to it and some of that works, but mostly all it is is Windows design principles hauled up by the roots and replanted in Linux. That's fine, but don't even call it superior design, or working software, or functionality incarnate. It's not, and there is no reason to seriously believe it will ever _fully_ live up to that sort of hype.
    It doesn't need to- somebody has to take care of the Microsofties come to Linux and looking for their taskbar and shortcuts. But put a cork in the KDE-uber-alles ranting, OK? You're flat wrong, precisely as wrong as Microsoft itself is when it makes THOSE SAME EXACT ARGUMENTS. And Windows, too, works- sort of. But I don't want to use it, either, and I don't.
  • He mentioned both license and looks. Your work-around takes care of looks, but not license. Linking something GPLd such as KDE with lesstif is actually legal, while linking KDE with Qt is not.
  • Posted by The guy in the next cubicle:

    Very Next-ish. Reminds me of the look and feel of the WindowMaker control center, which, half-assed functionality aside, I really like.

    Personally, I think GTK looks like crap (bring on the flames, I don't care) and I can't stand all of those stupid buttons that change color on mouseover. I feel like I'm looking at some "k-rad" Java applet on a badly designed Web page. And GNOME's panel is butt-ugly right now. They need to slim the thing down and do some major ergonomic work.

    Really, all of this pissing and moaning and flaming over how widget sets look is useless, and makes the freenix-using community look like a bunch of whiny teenagers. Yes, the QT license issue is a concern, but QT 2.0 will be OSS, and if people don't like the terms of use for KDE, they can use something else. Flwm and dfm together make a very nice (and FAST!) desktop. Bottom line, however, is that KDE is the only desktop I can give to my parents and expect them to use with ease. GNOME is not there yet.

  • How about FLTK [easysw.com]? YMMV regarding the visual attractiveness of it compared to the Big Two toolkits. I think there's also mention of other toolkits in their mailing-list archives.

    --

  • ...this page [free-soft.org], which lists (with links) nearly every toolkit known in this solar system. As for printing support or other solar systems, I plead my usual ignorance.

    --

  • Here's a tip: the majority of the AC posts are coming from a few clueless 14 year olds. If you think this is a forum, you're mistaken; it used to be a forum.

    TedC

    PS. Those KDE 1.1 Mac style menubars are cool!

  • Next thing I expect from you is a disgusting comment about masturbation.

    I have never made any references to masturbation on /. or any other online forum. I think you must be confusing me with someone else. I was merely commenting on the lack of substance in a large majority of AC posts -- your's would be a good example.

    TedC

  • The Gnome/KDE flamewars are getting rather.. old. If you like Gnome, use that. If you like KDE, use that. If you don't like either, don't use either one. If your idea of a good time involves baiting people or using massive amounts of four letter words in the description of the one you don't like, get a life.
  • I sure hope there will be a 1.1.1 maintenance release soon.
    I concur. There were a lot of pesky little crash bugs in the 1.0 programs that really got on my nerves. And, a really big crash bug (kfm) .. As far as I can tell, there were never any minor maintenance releases clear them up. Now would be a good time to start. :)

    This is especially important in light of recent release happenings (not just KDE - Linux 2.2.0, for instance). i.e., when complexity is such that indefinite freeze periods occur and the "during testing" pool of testers is limited. I think the Linux kernel strategy of responsively and constantly maintaining the stable tree is excellent.

  • Why not ask the supermarket for free food, because their "licenses" suck so much. If anything you're a complete ass for making such a big deal out of something that I doubt will have any effect on you, as I seriously doubt you've contributed any code.

    Before you attempt to discuss the GPL or the philosophies behind it, you should have a look at this document [gnu.org]. The moral opposition expressed there is not one against paying for software, it's against proprietary software.




    It's against QT's license to distribute it in a modified fashion. To apply one of the concepts it explains to your statement, if you get food from the store, the store no longer has the food when you leave. This doesn't apply to software.



    That said, I'm not a KDE basher. I actually use some KDE apps under WindowMaker. And, I contribute code to and maintain free software, so :P
  • Acutally, the GPL would -NOT- prevent commercial use of the free library. Quite the opposite. The GPL covers programmer's freedoms, not the price tag.
    (It's about Free Speech, NOT Free Beer.) Hence it would be perfectly possible and legitamate to sell GPLed software, PROVIDED the freedoms implicit in the licence are included.
    This means that the GPL CANNOT be used to inhibit the commerical selling of a product.
  • by jd ( 1658 )
    Then you'd get GCC/egcs for Windows, port Qt, and compile your package.
  • Every time I install a new version (because there is this new program I want to try out that need this new version), some of my other programs breaks.

    (a) Development libraries don't maintain backwards compatability.
    (b) If your programs are breaking, I'm afraid it's your own fault..multiple versions of GTK+ can live side-by-side quite peacably. (I have 1.0.6, 1.1.5, 1.1.6, 1.1.7, 1.1.8, 1.1.9, 1.1.10, 1.1.11, 1.1.13, and 1.1.14 installed) And all I did was "./autogen.sh && make && make install". (for the ones that weren't installed by my package manager).

    Clues are a recommended accessory when using development software.

    Daniel
  • I use linux since about 1.1.59, and wouldn't drop
    KDE for a fvwm setup unless someone pointed a gun to my head.
  • 1) The STL doesn't define a string class.
    Some STL implementations have one, but it is not a standard.

    2) STL containers can be used trivially with Qt.

    4) I have interfaced C++ libraries to Qt with no
    problem whatsoever (the mime++ class, specifically)

    5) Qt's QString class can now do unicode. How do you do that with the non-standard STL string?
  • I wrote one, so it must exist.
    And before anyone says otherwise: it was a semi-trivial task. 90% of the work was done automatically by a tiny python script.

    And no phony vtables were used, either.
  • -----------
    1)The STL doesn't define a string class. Some STL implementations have one, but it isnot a standard.
    --
    I think that you are mistaken, see the now accepted standard on strings. The standard
    most definately provides a string class and Qts is not to my knowledge complient.
    -----------
    The URL you provide contains a description for a string class. Yet it never says it complies to
    any standards.

    That description is part of a "working paper" dated december 1996.
    Yet, on my copy of egcs, which is dated in 1998, I see: "NOTE : This does NOT conform to the draft standard and is likely to change" on it's string header.

    Can you shed any light?

    --------------
    2)STL containers can be used trivially with Qt.
    I assume that you are only attempting trivial things with them then. Last time I read the
    Qt documentation, it did not provide a direct translation from types such as vector to
    their Array and such. Without such you will with suffer a penalty of having to convert
    types if you mix containers from Qt and STL. Although I admit I am hardly a Qt
    expert. If you are wrong, please send code.
    ------------

    Why mix them? If you want to use STL comtainers,
    use them. If you want to use Qt containers, use them. I see no contradiction between what I said and what you said, except for your trivial slam at my usage of STL.


    I must have deleted 3), sorry.

    ------------
    I have no knowledge of the mime++ class, so I won't comment.
    ------------

    You claimed interfacing Qt with other libraries was a problem. Can you give an example? A reason?

    ------------
    (about using wchar to do STL strings do unicode)
    ------------

    That is not really a solution. Or rather, it's a clunky, ugly, slow solution.
  • Well, I agree with you licensing-wise. However, it's worth trying things out to see why people like them. I hope the Gnome people and the KDE people at the very least use the competition for some time on their own desktops: cross-fertilisation is good. I'm actually waiting for GnuStep 0.6.0 due out later this month. Hopefully it'll be usable.
  • I think KDE is/will be the desktop environment for converted windoze users. However, that file dialog is a clone of the windoze file dialog - one more example of braindead design AFAIC.

    If you want to see how a real file dialog is done, take a look at X-file for OS/2 [bmtmicro.com].

  • SplitVT fixes that, anything else?
  • Saying "GPL should 'never' should be used for libraries" isn't a very accurate statement. LGPL -is- GPL with some exceptions for proprietary software linking.

    However, Troll releasing GPL would infact be closer to what they're doing, and drive free software even harrder, but not allowing proprietary software to be compiled for KDE...

    However, LGPL would - for the time being - be more desirble, to attract proprietary authors to GPL software. In the future, of course, all software should be GPL.
  • This brings up an interesting issue. OSS has been a great move for free software, making a definitive difference. Your point however, is a good one: let's not loose sight of GPL. OSS is good, [L]GPL is better. Distributing software between licenses isn't legal unless they have the same license, or a BSD license.

    That's why an LGPL library, like gtk+ is still more desirable than regular OSS - you can use the code in your own free projects. QPT'd code won't be very useful to anyone. Troll Tech needs to GPL Qt, since it would play such an integral part the system.

    A good rule of thumb, IMO, is to allow programs/applications/libraries to be OSS, so long as they aren't integral to the system. If they are, GPL/LGPL should be required.
  • Um, if i remember correctly, a LOT of GNOME is written by mexican coders.. yeah a lot of americans contribute, but the gut of it is Mexican! (I'm waiting for the taco module )
  • I really don't like the Motif look at all. Bleah. HP uses for everything in the HP-UX tools, and I think they've created the ugliest desktop (maybe 'cept for OpenLook) ever.

    I haven't decided between GNOME & KDE yet for looks, but let me make clear I respect the people who do both. I think Troll has been very reasonable meeting people's problems with the way Qt is liscensed.

    Eventually I'll prolly install both. I'll run whichever panel I eventually decide on, but I'll use apps from both. I can't live w/o KLyx, Iagno (Ian is renaming Gnothello) and Gnibbles (XNibbles gtk-ified).
  • Well, there are visual debuggers (ddd, xxgdb)
    which are actually pretty good but your statement
    "gcc has no ... debugger" causes me to want to
    ask "How much time have you actually spent using
    gdb?". Gdb is actually very nice and very powerful - it (believe it or not) has some features not found in any other debugger. I have
    used the visual c++ debugger and it is very good
    but I have never in my life found a piece of code
    I couldn't debug using whatever tools were available and frankly I've found that the biggest
    proponents of the visual c++ debugger were people
    who were didn't really want to do the work of thinking through the problem - they just wanted to
    point their magic gui at it and have the problem
    go away. The biggest impediment to debugging is
    deciding that one will actually do the possibly
    hard work that is required to mentally understand
    the problem and step through it to solve it. In
    summary if you haven't actually spent any time
    using gdb, reading its manual, learning what it
    can do you have no basis for criticizing it - though it is not inherently visual it really is quite good, I encourage you to actually try it before making disparaging remarks about its capabilities.
  • Well, there are visual debuggers (ddd, xxgdb) which are actually pretty good but your statement "gcc has no ... debugger" causes me to want to ask "How much time have you actually spent using gdb?". Gdb is actually very nice and very powerful - it (believe it or not) has some features not found in any other debugger.

    I have used the visual c++ debugger and it is very good but I have never in my life found a piece of code I couldn't debug using whatever tools were available and frankly I've found that the biggest proponents of the visual c++ debugger were people who were didn't really want to do the work of thinking through the problem - they just wanted to point their magic gui at it and have the problem go away. The biggest impediment to debugging is deciding that one will actually do the possibly hard work that is required to mentally understand the problem and step through it to solve it.

    In summary if you haven't actually spent any time using gdb, reading its manual, learning what it can do you have no basis for criticizing it - though it is not inherently visual it really is quite good, I encourage you to actually try it before making disparaging remarks about its capabilities.

  • I suppose everyone is entitled to their opinion
    but I have to tell you after seeing years of
    typical x apps and lots of gnome/gtk stuff I was
    pretty well blown away by kde. I have to wonder
    if some of this ranting is based upon religious
    rather than pragmatic reasons. I mean have you
    ever actually run kde? Though probably 8 of the
    12 platforms I have actually written commercial
    code for were unix variants I was always saddened
    by the relatively amateurish look of most of the
    x-based apps - netscape being one of the exceptions. I'm using kde now and you know what?
    Netscape is the least-spiffy looking thing on
    my desktop.
  • they could make money with a better license.

    __
  • by Kenelson ( 4445 )
    I am not sure about the canvas widget with printing support, but gnome canvas is the current work in that direction. The fact that DrawingArea is worthless for your printing usage is no surprise. It was meant for as a quick way for the user to add graphics without the overhead of writting a new widget. Comparing a canvas with a drawing area is misleading, as they have different purposes. You should compare a gnome canvas/print with the qt canvas.

    But neither of your arguments have anything to do with gtk--. Those are functions of gtk+ and not of the wrapper on top.

    There is nothing inherently a kludge in writting OO code in C and then wrapping it in C++. No one has ever cited a document showing that all wrappers must be bad. As gtk+ already has an excellent OO design very little changes were needed to make a good wrapper. Besides both Qt and Gtk-- wrap the OO X code in C so the point is mute. Both are wrappers of some sort.

    --Karl
    Gtk-- Contributor

  • by Kenelson ( 4445 )
    Despite your colorful descriptions there is a huge laundry lists of why it is a good idea to write an OO frame in a non-OO language. The most basic is freedom to design without the inherent restrictions of the OO framework the language provides. See the gnome FAQ if you want more.

    But this is beside the point as we are to be comparing a C++ toolkit with another C++ toolkit. Gtk-- interfaces to C++ completely including STL defined methods. Qt on the other hand, implements its own set of list, strings, and containers that are not nearly as well as the STL ones. Further, gtk-- works within the C++ defined framework to construct its signal system instead of building in with MOC a meta compiler. So honestly, gtk-- is more C++ than Qt. It you want to interface a C++ library to Qt, you will have even more cost that we do interfacing to C.

    If you honestly looked at gtk--, you would find it is a very clean toolkit class set. You can derive widget freely without worrying about the gtk+ internals, so what is your beef? And with our signal frame, your application will still be portable without a meta compiler.

    You will also discover that the extra layer accounts for all of one extra function call in 90% of the cases. This is hardly a huge overhead. If you read through the fine book More Effective C++, you will discover that C++ implementation of its own internal mechanisms is far higher than this.

    Both the printing needs and the gdk interface are properties of gtk and not just gtk--, so saying gtk-- sucks because you don't like the underlying gtk+ is being overly specific. It is like saying that you hate all Taurases when you mean you hate all Fords.

    I understand that we don't have the interfaces you require, but that functionality is the domain of gnome, not gtk. If you think the documentation is bad feel free to contribute some more usable docs. We welcome participation in our project.

    --Karl

  • by Kenelson ( 4445 )
    I am sorry that we don't provide a good printer interface, but that is as I say the domain on the gnome project. (I just wrap things) Asside from the lack of a printer interface, what else do you find deficent?

    I don't understand how you can think that we have some inherent restriction on our code. Yes providing a good wrapper is more difficult than writting from scratch, but we are not forcing you to work at the C level. We have taken the time and effort to completely cover it all. Any places that you must go and access the C understructures is a place we need to cover better, and we are happy to learn about such places, so we can repair them.

    If your advisors AI system was interfacable from C++ and you could not tell the difference from the interface or running it, than what difference did it really make. You are assuming that all wrappers must be bad and therefore gtk-- is bad. With that assumption, you could not possible accept that we have created a C++ library from a C framework.

    From the technical stand point, we did not just wrap the C gtk+ and call it good. (That was the gtk-- of about 9 months ago, it really did sucked.)

    Currently, our code:

    • takes over the gtk signal system
    • provides C++ equivelent mechanisms for parts that we found restrictive.
    • provides STL compatible access to the gtk structures.
    • takes over the callback system and transforms it to virtual functions.
    Following the next release, it will:
    • take over the memory management of the gtk-- structures that are linked to gtk+.
    • seperate into interface and utility functions.
    • increase the strength of our signal/slot mechanism far beyond that of Qt.
    • provide a translating mechanism that will transparently promote all C structures to C++ structures when accessed.
    Perhaps at that time you will reevaluate our design. Til then happy coding.

    --Karl

  • by Kenelson ( 4445 )
    Actually, we can add widgets to the C++ interface directly. Unfortunately then those widgets would not be available to the rest of gtk interface. We chose not to do more than wrap, because those interfaces are already being developed in C and it would be a waste of our effort to reinvent them.


    If at some time later gtk-- becomes wildly popular and C falls out of use, we can always port the gtk code directly to C++. However, as I am sure you will agree C++, has some diffiences that need too be addressed. Given the controversy, shouldn't we just wait for D. ;-)


    Good talking with you.


    --Karl

  • I am not sure what you mean when you say "gtk-- is such a piece of crap." Could you please be a little more specific as to what you find missing? Free software such as gtk-- only improves if we know what is wrong.

    Currently Gtk-- provides

    • A good signal system comparable with Qt.
    • C++ mechanisms to the gtk+ toolkit.
    • C++ access to gnome widgets.
    • A lowlevel interface to X drawing primatives.
    We have been making vast improvements in the documentation and the code readablity in the last two weeks. And we are preparing for a stable release to correspond with Gtk+ 1.2.

    Is there something more you desire, or is there some feature that we are missing that we should add? We would really like to know before we realease 1.0. Also when was the last time you looked over the interface?

    --Karl
    Gtk-- Contributor

  • How do you compile Linux with Visual C? This I would like to see. Actually, GCC existed long before Linux and was used from the start.

  • From my understanding, GNOME and KDE can compliment each other, and aren't really competitors.

    At least, that's how I seem to understand it.
  • For crying out loud lets end this myth that KDE is a memory hog.

    >man top

    Learn how to read the damn thing.

    Compared to KDE 1.0, KDE 1.1 is miserly!
  • ..I write a free software using QT. SOmething really useful. How can people, who use Windows benefit from it? QT is not free for Windows - so Troll is screwing people out of free software; while GTK is going to be ported and free to use. Such a shame gtk-- is such peace of crap.

    If you want, you can use the X version of Qt with an X server on Windows. Clunky, but it works.

    As for GTK--, it's not the only C++ wrapper for GTK; check out wxWindows, which wraps GTK, Win32 and others in a single class library.

  • Database access in an X toolkit? Yuck!
  • KDE is Kool, GNOME is Cool - I like competition :)

  • Yes, but it would deprive themselves from any benefits since they couldn't include patches into their commercial version (and after all, that's what commercial OpenSource licencing is all about).

    A GPLed FreeQt would force an irreversible fork because the GPL is by far more restrictive than the QPL: Once you're in there, there's no way out again, while the QPL allows free contributions to remain free (you don't have to hand over your copyright, source has to remain open, etc.) but in addition to that allows commercial relicencing, which is essential if we want Linux to become the mainstream desktop OS.
  • Well, I want a coherent desktop environment, but not at the price of system stability and usability. I guess I want to have my cake and eat it too. Is that too much to ask?!

    I have not tried any recent GNOME offerings, but I did install KDE 1.1pre2. I only have 32MB of RAM and 18MB of swap. Yes, I know, my fault for setting it up wrong so long ago. Linux needs something like HPUX's logical volume manager so I can just add to the swap.

    Anyway, I run KDE, start up Netscape, Gimp and two of those Kvt terminals. Those stupid terminals take up 4MB of RAM. What the hell?! Well, as you might guess, my system starts swapping like there is no tomorrow. It becomes completely useless and I have to hit the reset button :(. Well, after that experience it was back to FVWM for me! I appreciate a window manager that starts up in about 5 seconds ;)

  • I don't know what you're talking about wrt KDE's stability. I've been running it since Beta1, and now have gotten up to 1.1pre2, and will be downloading v1.1 tonight. In this time (over a year now) I have seen not only it's feature list, but it's stability grow with every release. I can personally attest to the service it has provided me. I've been through at least fifteen different window managers, WindowMaker being my favorite. I've even tried to compile the last couple (two, maybe three) point release of gnome, and it has never worked. KDE has compiled and installed out of the box every time, and as run with no problems. Political viewpoints aside, KDE provides so much more to the end user, with so much less of a hassle. I have to upgrade so much stuff before I can even hope to run the new gnome point release it's not even funny. And I have a RedHat 5.0 system, an install that's not even six months old. Tell me somewthing isn't wrong with this picture.

    That said, I cheer on the KDE folks whenever I can. I have nothing against gnome, other than the fact that it requires about a billion times more _Stuff_ (libraries/etc) to run and has never completed a compile on my machine. I wish their project well, but KDE is here today.

    You may make fun of Microsoft's "Where do you want to go today?" campaign, but in the long run that's all that really matters. We have a working solution today with KDE, so why not take advantage of that? If gnome ends up being somehow superior, I am quite sure it will take it's place above KDE - but that day isn't today.
  • I've been using a recent snapshot for a while and it's extremely stable. I've only used a fraction of the available KDE applications and they've been working great too. KFM has yet to crash on me (and I've used it to browse slashdot too) so no complaints here. In fact, I like 1.1 so much that I've made it my main desktop and WM whereas before I was still using Fvwm and FvwmModules and the occasional KDE app.

    So although I hear there are some bugs that went out in 1.1, it should actually be pretty stable and 1.1.1 has already been confirmed anyway.

  • >Personally, I think GTK looks like crap [snip]...
    >and I can't stand all of those stupid buttons
    >[snip]... And GNOME's panel is butt-ugly [snip]

    >Really, all of this pissing and moaning and
    >flaming over how widget sets look is useless...

    Oh really? You're sure doing enough of it, asshole.
  • As many of you here know, my company uses both GNOME and KDE on desktops. Both have their unique featuresets. We do alot of development in Ada95 on Linux building very robust GUI apps. Because GNOME has Ada bindings that are very workable, GNOME is a no brainer. The GNAT compiler works on Solaris, HP-UX, SCO etc. and is not just Linux centeric.
  • Has anybody actually run any tests comparing the latest incarnations of VC, EGCS, and PGCC? That would be interesting to see..

    Also, on the original issue: I actually do think that having a more or less stable and *documented* library such as QT helps. The docs for GTK are unfortunately sketchy at this stage.

    Just my $2E-10 worth...

    Marciano
  • Why preannounce it, just like LinuxToday?
    It's true that it'll be put on ftp.kde.org very soon, but only the directory structure is present!! Nothing's there now!
    Everything is still under the unstable branch. Period.

    Maddog
  • And people like you who seem to hide behind AC just to bitch and moan everytime people aren't jizzing their pants over Linux need to get a life. 'Nuff said.
  • And you're someone I'd want to associate Linux with, because you and people like you give me a damn good excuse to bash Linux.
  • You must have missed that ;-)

    Who said he was alluding to a monitor?
  • Ada is a language. The Ada compiler (that I know of) that is supported by egcs, Gnat, works primarily with Linux, and a few other comercial Unicies. Chances are if you're as Linux-centric as Gnome, you'll have Ada bindings, otherwise I doubt it.
  • Yeah you must speak the gospel truth, with out a doubt *nod*. Hey I've got one, if you're not a complete zealot, like moderation, functionality over flash, use KDE.
  • That's such a load of shit. If what you're spewing is really worth saying, wait until you're home from work. If you don't have a computer or inet access at home, go to a public library. There's no reason to be an AC, esp. because you're not required to give out your true email address to all of /.. Get a life.
  • So start work from the C bindings that Roberto made if you're so hell bent on a C toolkit. Why not have a punch card toolkit?
  • Yeah and the nice jump from 0.33 to 0.90 with no coresponding jump in stability.
  • Well, that said, I really don't agree with you there. Think about it, how is TT going to support Qt if there's a bunch of different modified versions out there. Look at the Linux kernel, each distro has its own kernel pretty much. That's fine because there are rather large companies supporting them, but TT are the only people supporting Qt right now, and such high level projects tend to attract more script kiddies, meaning I seriously doubt that one big company would take over the support... and TT's "support" is bad enough as-is.

    That said, I use Gimp, wmaker so there.
  • Uhh no. A quick check on the master site reveals a Linux binary (for libc5 and one for libc6), a singular binary for Sparc Solaris 2.5.1 (at least two Solaris versions behind..), one for an old version of HP-UX, one for an old version of Irix, DU, and an old version of Solaris x86. Either way, it's somewhat encompasing, but as I said, it supports one free Unixish OS, and some outdated comercial ones. Not good enough. As for the non Unix OSes, I'm not really concerned with those :}
  • Oh back that up please. Your unfounded comments are on par with the average /. reader.
  • Now I'm curious, what if any documentation on Qt themeability is there? That would make porting to Qt 2.0 so much more fun :}

    - alex
  • Each distribution pretty much tweaks the kernel config to their own needs. So each distro ships with a different set of drivers compiled in. Espero que entiendas (I think that's what I meant to say).

    Debian provides support, no? And yes I remember Debian for that extremist Bruce? Perens? What a good reason not to use Linux.
  • Solaris 2.6 and 2.5.1 are oldish for instance. Sure Sun is good at supporting them, but they're still old, and 2.5.1 is just awful. I don't know much about HP-UX versioning but I saw binaries for only one version (10.20), which IIRC is not the most current. Sure there may be quite an installed base of poeple with Solaris 2.5.1, but the sheer fact that there aren't binaries for more OSes (or versions thereof) proves my point. Ada is a lesser used language, it doesn't run everywhere, it won't run on my FreeBSD box, etc. I'm sure Ada bindings for KDE/Qt would be a great asset, but seeing as KDE doesn't focus on one OS or one version of an OS, I don't think that Ada bindings will happen as soon as they did for Gtk+ which is very tied in to Linux and Linux only.
  • Uh no. Gnome runs on FreeBSD, but you've got to use the ports. KDE runs well on FreeBSD, but you generally should stay away from the KDE and Qt ports IMO.

    What really shows why I don't like Gnome is its configure stuff. It's really bastardized the autoconf mess. If you want to run code out of Gnome's CVS tree, chances are it won't run on FreeBSD (or run period..) without a lot of tweaking. KDE tends to be easier to get up and running from the CVS tree.
  • Tacos son comidas americanas. Por favor piensa antes de hablar.

    (See how dangerous High School Spanish classes are?)
  • I don't know about the various Gnome apps.. I think GNUmeric reads/writes Execl files.. I know the KOffice programs all use XML.
  • Uh no. KDE 1.2 has been shipped with a few known bugs. Me.. I don't like this policy of meet the deadline rather than meet quality standards. But some of the powers-that-be just couldn't wait. Not much can be done about that. And yes I hope 1.1.1 is shipped soon.
  • You obviously haven't looked at it yourself, otherwise you wouldn't make such factless stupid comments. There are bindings for C, Perl, Python and perhaps Tk for KDE.

    And you're obviously not sorry to say what you've said. So stop saying you're sorry.. in the words of a wise person I once knew.. you're cheapening the words [dumbass].

    I'm sure if there was a demand for Ada bindings, people would make them. Hell I'd volunter to make them, but it means I would have to ditch FreeBSD, something I'm not willing to do (well that and I'd have to learn Ada..).
  • ..just installed 1.1pre2....

    oh well,
    will wait for the qt-2.0 based one now...
    (scrolling wheel my baby.. love it)
  • ..I write a free software using QT. SOmething really useful. How can people, who use Windows benefit from it? QT is not free for Windows - so Troll is screwing people out of free software; while GTK is going to be ported and free to use. Such a shame gtk-- is such peace of crap.
  • Are you leaving on other planet? People have to had there work done, and the platform they use for that is not usually there choice. I have to use NT often; I really fucked our computer support to allow to get a Linux box on network. Be real and get you head out of your butt. gees.
  • Just take a look, for example
    at this [troll.no]
  • ..by the license. It says X Windows system only. Why a distinction? I understand having to pay for commercial development.
    You see, they effectively killed the Harmony project, which would have allowed for cross platform development, if under GPL and if ported. Now nobody will ever write a good toolkit, that can be made cross-platform. Bastards.
  • ...would you like to bind C++ to C? I thought the progress was moving in a direction from procedural to object-oriented programming.
    What C++ lacks that C has? (asctually, IMO, that's the probel with C++, but tastes aside).
    Compilers for some DSP? Are you using a window toolkit on a DSP?

    Graphical tookit have to be object oriented, sorry, that's the only sane way to include asyncronyous, random user generated events.
    Of course it does not mean you have to use an OO language,- hell, you can find a way to code it in assembler,- but that (OO) surely does help.
    In any case, for anything resembling a good UI you will have to expand the toolkit, to match your needs and to create your own compund objects. It is to be done in the same language the toolkit is written in. And that's better be an OO language if you value your time. Just inherit from widgets you like, and create what you need and give it to other people to use. I do not even use a single default Component in Java or Widget in qt, only my own library, that does what I need. And you can use it too. What a pain in the ass to try to use C for such coding.


  • ..would you create a compound widget with a clean interface for future reuse in C? With a procedural language you are limited (bound, if you wish) to a toolkit, which is very limiting. What if you want a Plot widget? A systemStatusMonitor widget that includes a plot widget? Now change it to a different type of plot.. so on. How would you use exeption handling? You just have to implemnt all OO functionality crawling on your knees. Ouch.
    I agree C++ is not the best. Personally, I am waiting for a good native before time compiler for Java (hello Cygnus :)

    So you propose to put development effort into crappy hack of a toolkit, just because someone doesn't want to learn C++? What a waste of time.
  • GTK+ is fine. I looked at it. If you just use what is there. But I need custom widgets. That behave like an object. Making it happen in C is a pain. It is not an OO language. Yes, I can write a wrapper, but why not just to start from a good approach?
  • comparing C++ and C speed under EGCS

    was this [stanford.edu]

    Seems you are wrong..

If a thing's worth doing, it is worth doing badly. -- G.K. Chesterton

Working...