This discussion has been archived.
No new comments can be posted.
Interview with theKompany.com's Shawn Gordon
Related Links Top of the: day, week, month.
As you will see, I told them, in no uncertain terms, to see Figure one. -- Dave "First Strike" Pare
Embedded/Zaurus software (Score:3, Insightful)
Re:Embedded/Zaurus software (Score:2)
Re:Embedded/Zaurus software (Score:5, Interesting)
He has consistantly listened to ideas from his customers and has discussed the feasability of each option.
There was a nice healthy discussion regarding distributing Kapital as a statically linked application recently. While I didn't like his answer or agree with the end result, he DID participate and give his reasoning in a clear manner and with extreme consideration to the customer.
Re:Embedded/Zaurus software (Score:2, Funny)
and even on Christmas eve, Shawn was responding to messages on the mailing list
Maybe he's Jewish?
Second best? (Score:4, Interesting)
So, my question is, why is KDE considered second best? Are there technical reasons, or political, or what?
Re:Second best? (Score:5, Informative)
[sigh] It's not. Look, I'll try and tell it like it is, but nobody is entirely neutral in this debate OK?
Basically, with Mandrake looking like it's going down the tubes, there are 2 major commercial Linux desktop distros left, Redhat and SuSE. There are others like Xandros of course, but they are more focussed on providing an "appliance" style OS, rather than staying level with the current cutting edge in Linux development.
RedHat are popular in the states, and are "biased" towards Gnome, that is they have more Gnome hackers with experience than KDE hackers. As such, their distro focusses on Gnome more than KDE. SuSE is similar but opposite, they focus more on KDE than Gnome and afaik don't have any Gnome hackers on the team.
When Redhat 8 came out, as I'm sure you noticed, they attempted to equalize the desktops somewhat. BlueCurve was an attempt to give Redhat a distinctive brand on the desktop and it worked tremendously well. Nonetheless, some people involved with KDE got a bit upset, because KDE has its own brand (as does gnome) and Bluecurve changed that.
Today the desktops are basically equal, although they are stronger in different areas. So, GNOME has better usability IMHO, but KDE has more features. I should think theKompany likes KDE/Qt as a developer platform more because Qt is commercially supported, has professional docs and is more cross platform, so (if they pay) they can sell their apps on Windows and MacOS as well. Of course he has hackers with KDE/Qt experience which also tips him. On the other hand, GTK is more Linux specific, but has some cooler features. Some people will tell you that GTK is harder to program for, but in reality that's not the case, if C++ is your thing then both Qt and GTKmm are excellent.
I think you're exaggurating when you say KDE is slower and uglier on redhat. I think the BlueCurve artwork is great, but you can always retheme it easily, and it should be no slower.
Re:Second best? (Score:3, Insightful)
> stronger in different areas. So, GNOME has better usability
> IMHO, but KDE has more features.
This is a shortsighted opinion of yours and doesn't reflect the reality and obviously show that you never tried KDE for longer than 5 minutes. I know both Desktops on it's best using GNOME CVS HEAD and KDE CVS HEAD here. In my opinion GNOME hangs behind KDE in many ways. I don't want to make GNOME look bad now but I want to give you a fair objective view of the facts.
There are a lot of issues within GNOME such as integration, usability, consitency over the dialogs etc.
GNOME is not as integrated as KDE is. Have you ever dealt with KDE and seen how all the apps are playing together ?
Usability, have you seen what happened recently to GNOME's interface it toally alienate GNOME from the rest of available applications that work and require X with any toolkit.
Consistency. We all know that with the ongoing versions of GNOME the developers are trying to get consistency into the applications but yet all dialogs are still looking differently. There is no real pixel exact layout of the dialogs, menus and windows because many GNOME applications are written either with the GUI hardcoded, or Glade or Bonobo therefore all UI's are looking differently and makes it hard for the UI reviewer to get a structure inside.
On the otherhand KDE's ui are mostly external in separate *.ui files which makes it easier for the UI reviewers to concentrate on the UI and pixel exact layouts.
That's only a minor point of the whole mountain. KParts for example, every new written app follows a strict rule of KDE and embedds itself seamingless into the whole desktop. It has a Webbrowser embedded in a Filemanager and both of them are really usable and and and..
I could write half a day and name more and more features and pros which makes imo KDE far superior. KDE as is right now are milies if not lightyears ahead of GNOME but it's understandable that GNOME people don't want to have their desktop look bad but face it as is GNOME has a hard and long way to go to at least keep up with a little bit of KDE's features. Keeping up with KDE itself is a lost battle.
Re:Second best? (Score:2)
After all, what better way to get a reasonable judgement on how consistent and usable a software package is than to use CVS builds?
GNOME is not as integrated as KDE is. Have you ever dealt with KDE and seen how all the apps are playing together ?
I wish people would stop calling this "integration". You can have apps that conform to a common IPC standard without "integrating" bits and pieces of them into each other, a la Explorer, Office, and Windows.
Usability, have you seen what happened recently to GNOME's interface it toally alienate GNOME from the rest of available applications that work and require X with any toolkit.
No. What are you talking about?
Consistency. We all know that with the ongoing versions of GNOME the developers are trying to get consistency into the applications but yet all dialogs are still looking differently.
True. The same is also same of KDE apps. Oh, maybe not the very core ones, but there are only a few of those.
There is no real pixel exact layout of the dialogs, menus and windows because many GNOME applications are written either with the GUI hardcoded
I've done this, once, to learn the toolkit reasonably well. It's quite time-consuming, and not something most people are going to want to do.
or Glade or Bonobo therefore all UI's are looking differently and makes it hard for the UI reviewer to get a structure inside.
Um...what? Glade is a *good* thing for UI design! It makes it easy to *fix* inconsistencies!
You mean high-level widgets, like Open and Save dialogs? Gnome provides standard versions of those, exactly as you want.
That's only a minor point of the whole mountain. KParts for example, every new written app follows a strict rule of KDE and embedds itself seamingless into the whole desktop.
Errr...that's fluff. It doesn't say anything.
It has a Webbrowser embedded in a Filemanager
"Microsoft did it, so it *must* be a good idea!"
GNOME has a hard and long way to go to at least keep up with a little bit of KDE's features. Keeping up with KDE itself is a lost battle.
I know one KDE fan in person. He uses GNOME/GTK apps on a regular basis, because the KDE application set size doesn't cover all his needs.
I don't know any non-KDE users that use KDE/Qt apps (with the possible exception of licq -- its Qt plugin is somewhat nicer than its gtk plugin, though if you wanted to avoid Qt by usin the gtk plugin, you certainly could).
Re:Second best? (Score:3, Informative)
SuSE have at least one Gnome hacker on the team, but they have certainly been more focused on KDE. For instance Waldo Bastion who I think still leads KDE development is a SuSE employee.
"Today the desktops are basically equal, although they are stronger in different areas. So, GNOME has better usability IMHO, but KDE has more features. I should think theKompany likes KDE/Qt as a developer platform more because Qt is commercially supported, has professional docs and is more cross platform, so (if they pay) they can sell their apps on Windows and MacOS as well. Of course he has hackers with KDE/Qt experience which also tips him. On the other hand, GTK is more Linux specific, but has some cooler features. Some people will tell you that GTK is harder to program for, but in reality that's not the case, if C++ is your thing then both Qt and GTKmm are excellent"
I liked to use both desktops
Re:Second best? (Score:3, Insightful)
You really should do a bit of research before spouting off like this. Mandrake is far from going down the tubes if by saying "going down the tubes" you mean going out of business. They had some problems, they have since fixed those problems, but are having some issues with a lack of available cash on hand to cover their debt. Due to that, they needed bankruptcy protection while they raise more cash and continue working on the next release of their OS and other products.
They are far from the point of throwing in the towel and closing their doors. I think a lot of people see this as the end of Mandrake because so many other Linux companies have gone under over the past few years. That does not mean that Mandrake will not be able to make it. They have a very popular distribution that is #1 in some markets and #2 to RedHat in most other markets.
It's a tough time for most companies right now. Not only is the US economy depressed, other economies around the globe are struggling as well. This has an effect on lots of business and to think that a few temporary setbacks are enough to end a great company such as Mandrake... that's nuts. I'm not claiming that Mandrake will be around forever. They may eventually buckle and fold but that won't happen any time soon.
I think it's worse than you make out (Score:2)
Um...
They had some problems, they have since fixed those problems, but are having some issues with a lack of available cash on hand to cover their debt. Due to that, they needed bankruptcy protection while they raise more cash and continue working on the next release of their OS and other products.
Ah. You *do* realize that companies that are going to go out of business act as if nothing is wrong up until the last possible minute to avoid losing customers, investors, and credit? "Lack of available cash" and asking for donations sounds to *me* like serious trouble.
I'm betting against Mandrake existing as a company in three years, but we'll see.
This has an effect on lots of business and to think that a few temporary setbacks are enough to end a great company such as Mandrake... that's nuts. I'm not claiming that Mandrake will be around forever. They may eventually buckle and fold but that won't happen any time soon.
Happened to Loki. Can definitely happen to Mandrake.
Re:Second best? (Score:2)
Have you actually ever used gtkmm? The only people I've known who ever liked it were complete C++ purists who hated the idea of moc, or people who've never used Qt.
Yup (Score:2)
AFAI can tell, SuSE is in even worse shape than Mandrake, at least in the States. Red Hat is the only one that's doing particularly well. I think their approach was better -- focus on the most lucrative markets quickly (business server stuff) to build profitability, then use this established foothold to start going after the desktop market. Mandrake (and to a lesser extent SuSE) tried to be pure desktop distributions at a time when Linux really wasn't all that competitive with Windows for the average user.
RedHat are popular in the states, and are "biased" towards Gnome, that is they have more Gnome hackers with experience than KDE hackers.
Yup. I could never figure out why the KDE folks can complain so bitterly about RH not throwing money at beefing up their KDE support. I don't see the GNOME folks complaining about Mandrake not supporting *them* well.
Some people will tell you that GTK is harder to program for, but in reality that's not the case, if C++ is your thing then both Qt and GTKmm are excellent.
MMmmm...I used GTKmm for a while, and while IMHO it fits better with the OO design of C++, it's really easier to use if you're familiar with GTK+ than if you're just starting from scratch. Incidently, Guillimue (I'm sure I mispelled his name) Laurent, one of the GTKmm designers, got fed up with how difficult it was to cleanly do a GNOME C++ API (unlike GTK), and ended up moving to Qt.
I think you're exaggurating when you say KDE is slower and uglier on redhat.
I've found KDE to be slower than GTK, period.
Uglier is, of course, a matter of taste.
I think the BlueCurve artwork is great, but you can always retheme it easily, and it should be no slower.
I'm not a huge fan of BlueCurve, but then I like a rather spartan interface...
Fine-grained packaging (Score:2)
I can compile and install all KDE in 1 day (in a PIII), with only 3 packages needed for the others.
I'd be more inclined to say that finely-grained packaging is a better idea than monolithic packaging.
There are large parts of GNOME that I'm don't use (Nautilus, Control Center, the Panel). I simply don't install those, and use the chunks that I want. With KDE, you're stuck with all kinds of crap that you may or may not want. RPM *should* be finely-grained.
Second, yes, it's painful to manually download and build each package. Which is why we have things like emerge and apt-get:
APT 0.5.4 implements a new command: apt-get build-dep. This command will try to retrieve every build dependency of a given source package. Support in APT-RPM was already implemented, which turns APT into a fantastic option for users of unsupported ports (like Conectiva for PowerPC), and general package building tasks.
Don't blame *GNOME* because they packaged things the way RPMs should be packaged and then you chose not to use the right tool to build them. That's completely unfair to them.
Re:Second best? (Score:3, Informative)
There IS an option not to GPL the application, but that requires paying licence fees to Trolltech, the company that made QT.
Commericial companies get wary of these issues, since they would always like to preserve the option of selling the software, or at least reserving their rights from their code.
Sun, from their evaluation report, seemed to have chosen GNOME as their future desktop mainly because of the licensing issues, but not of technical merits
Re:Second best? (Score:2)
It was probably also because they could easily hire a repository of experienced hackers in the form of Ximian, for which there is no such equivalent with KDE (theKompany simply writes some apps that work on KDE, they don't actually contribute much to the project itself iirc).
Re:Second best? (Score:2)
What advancements Linux brings, eh?
Re:Second best? (Score:2, Funny)
Political. KDE uses a GPL'd library AND C++, therefor its bad. Using anything more advanced and powerful than C is an unforgivable sin in the "open source" community.
Re:Second best? (Score:2)
Oh well... I just installed RH8 a couple of weeks back and I love the smooth interface. It really rocks in 1600x1200 once you've made the sidebar "tiny" at the bottom. I also like the little checkbox/exclamation mark applet which tells me when updates are out without me having to bother looking myself. That's probably available in KDE too though.
Bottom line is: use what you like/want. Program for what you like/want. Variety is good.
Daniel
"Race KDE cannot win" (Score:5, Insightful)
KDE people also have the weird habit of producing their own versions of various pieces of software. Surely a conservative decisionmaker will choose a desktop-agnostic Mozilla or OpenOffice over the KDE-specific versions. KDE applications might do better by just dropping the K from their names, thus competing on their own terms (snappines and other virtues associated with Qt).
Note that I have been KDE user in the past (alternating with less popular lightweight wm's), but Gnome seems to finally have gotten their stuff together with gnome2.
Re:"Race KDE cannot win" (Score:5, Informative)
> a desktop-agnostic Mozilla or OpenOffice over
> the KDE-specific versions.
You mean like those people at Apple who chose KHTML for Safari?
Re:"Race KDE cannot win" (Score:2, Troll)
Re:"Race KDE cannot win" (Score:2)
Re:"Race KDE cannot win" (Score:2)
Re:"Race KDE cannot win" (Score:2)
Very, very few people on Linux use KHTML (according to my website stats at any rate). Gecko rules them all, with IE coming in at second place, probably due to people visiting at work, or curious Windows users.
Apple seem to have chosen KHTML purely on the basis that when they made the decision it was faster than Gecko (actually since then Gecko has been speeded up by something like 20% and is closing to catching KHTML while keeping its featureset).
I don't buy the idea that they chose it because it was easier to screw around in the internals - Gecko is such a strong engine already that they wouldn't have needed to do the kind of deep surgery they had to do with KHTML. I think basically they knew when they started that by the time it came out Macs would be being kicked for being slow, so they optimized for speed from the get go.
So the decision by Redhat to use Mozilla instead of Konqueror makes perfect sense, considering that most end users won't care how "integrated" Konq might be, Mozilla renders the most pages and has tabs. And these days it's pretty fast too, so really I think the Apple/Safari thing is a bad example to choose.
Re:"Race KDE cannot win" (Score:3, Insightful)
I love KDE, but loath the license. I think all GNOME people are fans of KDE, they just won't come out of the closet because they are scared of the license.
What that the License gets you though, is the ability to ship the same high-grade apps on Windows as on Linux, Mac OS X, and whatever other platform you want. This could singlehandedly be the missing element to bring Linux to the masses. If they use cross-platform apps (Kapital for windows), then switching out the OS is small potatoes (Kapital for Linux) because there are no file-format lock-in issues.
I blame several people. KDE, and GNOME are all to blame. If GNOME people worked on Harmony, they'd be set Free, and there'd be no lisensing issues. KDE was foolish for choosing a toolkit with such a license (but it is a GREAT product, regardless).
Qt are the big winners, yet they have an excellent product and deserve to be commended.
BRING BACK HARMONY. Ironically, that will bring harmony to the Linux world. (Harmony is a GPL replacement toolkit that as started then stopped when the Qt license got more OSS friendly)
Re:"Race KDE cannot win" (Score:3, Insightful)
That's very true, but a straight C API is much easier to wrap in an OO framework. OTOH, a C++ API like KDE is a PITA to bind to other languages, both OO and other. Just look at the gyrations Kylix has to go through to bind one OO framework (CLX) to another (Qt). Unless there is one universal OO mechanism (such as
This is not to say that I don't like KDE and Qt. I think Qt is a great example of the benefits of a well designed OO framework. I'm merely pointing out the hassle if you happen to not be a C++ fan.
Where have you been for the last two years? (Score:4, Insightful)
Re:Where have you been for the last two years? (Score:3, Informative)
Re:Where have you been for the last two years? (Score:3, Insightful)
Hmm, well I may be somewhat redundant, but I've been looking for and evaluating several cross platform GUI toolkits.
I started with QT, because a few years ago, did some work in it. Doesn't look like it's changed much, but to write apps that run on Windows/*nix, I've gotta cough up $3000 USD per seat.
Then I grabbed GTK+, wrote a simple app, used glade to make the GUI, and compiled on Linux and Windows -- great says I. Then I wondered why the Windows version looked like the Linux version. Strange, windows programs should look like Windows programs IMHO
Then I found wxWindows -- this is to me the one that's had 10 years of development, yet next to no press. They do what NO ONE in the GUI toolkit business has thought of -- the apps look like native apps, yet the same source code compiles and runs on multiple platforms ! That and it's LGPL'd so I can make OSS stuff, closed source stuff, etc.
Another plus I find is that unlike most OSS stuff that's made for Windows developers this thing supports (very well) Visual C++, and Borland C++ builder.
One thing that's a complete pain in the neck to me is building stuff with cygwin. If I'm programming on Windows, I'm going to have a decent development tool for the environment, and sorry gcc isn't it (it's great on *nix, but that's because the OS has a good shell already).
From looking at the documentation and investigating a little, it looks like I can take my MFC programs and port them (the parallels are amazing -- and the API is probably a lot better than MFC from what I've read).
I think this is the ultimate solution for me, I can write apps that run on Windows, *nix (wrapping either GTK+ or Motif/Lesstif), and MacOS 9 and 10.
If the native OS doesn't have a widget, wxWindows emulates it (such as MDI on GTK+).
Re:Where have you been for the last two years? (Score:2)
If it has satisfactory performance at this point, ship it. Otherwise, port it over to straight C++ wxWindows.
Re:Where have you been for the last two years? (Score:2)
Re:Where have you been for the last two years? (Score:4, Informative)
Two major problems:
1) It means you can only write GPLd software with it. Apple were pretty keen not to let Safari be open sourced, though god knows why, it's hardly a cutting edge browser so they actually had to produce Qt wrappers and de-Qt parts to prevent it becoming GPL software.
2) It's only free software when using X11. That means KDE software can't be ported to Windows or MacOS and use the native graphics layer on each, they have to use an X server. Also, if one day (unlikely but possible) we all decide to move away from X, unless Trolltech update the license KDE is kind of shafted. Obviously that's the least likely of all scenarios, and anyway I expect Trolltech would just update the license, but you get the idea.
Of course you can write commercial or portable software with Qt, but have you seen the prices? It's $1500 per developer for the professional edition, meaning unless you are a rich company you have to make your software GPLd and X specific (which is what they intended, but hey).
In contrast, GTK can be used for commercial and portable apps, and really it's quite a good toolkit these days. Hence the flamewars.
Re:Where have you been for the last two years? (Score:2)
I don't know all the details on the license bundled for this package, but I saw a boxed version of QT on Trolltech's website the other day : http://www.trolltech.com/products/box.html [trolltech.com]
The page says "The Qt 3.0 box is offered in addition to the Qt license purchase for $149". Not sure, but it seems that you can use it on Linux/Win. Anyone know more about that, especially the license (can you write non-GPL apps (LGPL, BSD, closed, etc.) ?
Re:Where have you been for the last two years? (Score:2)
In other words, after paying $1,550 for a license and $480 for support, you have the option of paying an additional $149 for media.
Re:Where have you been for the last two years? (Score:3, Insightful)
That almost certainly is not why Apple developed KWQ (the abstraction layer that lets them tie Safari into Cocoa). They did that because Apple's own browser needs to have a good, thorough implementation of the Aqua user experience. Apple cares a lot more about the user experience (and thus is willing to go to the extra work to have it) than the typical
Re:Where have you been for the last two years? (Score:2)
Silly? And what standards do you know of that are based on GPLed softward? Hell, I don't think the internet would be using TCP/IP (it might not even exist) if it wasn't for the Open Sourcing of BSD and it's TCP/IP stack.
How about SSH? IPSec? Anything that has caught on has had a BSD-licensed implimentation that developers were free to use in any way they wanted.
If you think that any GPLed software is going to become a standard, you are very naieve, or have been taking rms's statements to heart too much.
Re:"Race KDE cannot win" (Score:4, Interesting)
This FUD should have been dispelled a long time ago. The C++ bindings for GTK/GNOME are excellent, in fact they are mor C++ish than Qt, as they make proper use of the standard library, as opposed to Qt that reinvents a lot of it (QString et al) for portability to ancient platforms and needs stuff like a preprocessor for its object model. Check it out [gtkmm.org] before you bash gnome again.
If you like C++, use it! Nobody is forcing you to use C. The fact that a lot of Gnome software is written in C is because the coders prefer C, that's it. No, really. They do. C++ is a hard, amazingly complex language that isn't to everybodies taste.
Look at Windows. MFC is just C++ classes aound C stuff.
The MFC is a good example of how NOT to make bindings, don't write off all language bindings because you had a bad experience with one.
What that the License gets you though, is the ability to ship the same high-grade apps on Windows as on Linux, Mac OS X, and whatever other platform you want.
Only if you stick to Qt of course, and pay up (a LOT of money) for each developer. If you want to use the KDE classes, not all of them are available on Windows or MacOS, so....
This could singlehandedly be the missing element to bring Linux to the masses.
Actually GTK apps are ported to Windows far more often, because you don't have to pay to do so. Try again.
BRING BACK HARMONY.
Oh yeah, that's nice. Why not, and destroy TrollTech at the same time. Do you have any idea how much effort would be required to recreate Qt? A widget toolkit is often many millions of lines of code, and Qt doesn't just do widgets, it does strings, threads and lots more.
The solution is to make GTK and Qt interoperate better, share theming engines and so on, then you can choose which toolkit you prefer. GTKmm and Qt are basically very similar anyway.
Re:"Race KDE cannot win" (Score:3, Informative)
FWIW, FLTK [fltk.org] provides the same advantages as Qt but is provided under the LGPL and is a lot smaller and faster.
The 2.0 release (currently in development) will take us beyond what most of the current toolkits can do, and is the basis of the Equinox Desktop Environment [sf.net]. At the same time, 2.0 retains the same goal of small size, high speed, and portability/cross-platform programming.
Re:"Race KDE cannot win" (Score:2)
The screenshots are fairly horrific looking. The only things that look half-decent are the gnome icons I came across.
It may be superior, but ugliness goes straight to bone (in my case at least). If I recall, I was so disgusted with the QT "digital number" widget that I had a hard time seeing the better parts of QT. It still haunts me; "The horrrroorrrrrrr!!!!!"
Outdated opinion (was: Re:"Race KDE cannot win") (Score:5, Informative)
BRING BACK HARMONY. (....) Harmony is a GPL replacement toolkit ....
I'm getting the feeling that the poster doesn't know that QT now comes under the GPL license (instead of their original non-GPL one). The poster loves KDE, but loaths it's licence. The license is GPL! Nothing wrong with that (as the poster likes linux).
Funny (or not) thing is that RMS is basically backing gnome (LGPL) while at the same time advising everyone to stay clear of the LGPL because it is an inferior language. GPL is preferred. KDE&QT is GPL, so it would be better to put the FSF's weight more (and more openly) behind KDE. By "better" I mean from a purely FSF-political standpoint.
Reinout
Re:"Race KDE cannot win" (Score:2)
Re:"Race KDE cannot win" (Score:2)
> Considering that so many of the leading KDE developers worked for Trolltech, this didn't look like it would ever happen.
At that time, very few did.
C is for interfaces (Score:3, Insightful)
Fair enough. Window environment programming really does benefit from a language with solid object oriented programming support.
But, C++ is not well suited to interacting with other languages or truly dynamic library use. These are two key elements of anything you're planning on making "core system libraries". Personally I include my GUI libraries in the fuzzy mess I label the core system libraries.
C's interface is the defacto interlanguage interface. Most mature languages develop some way of relatively easily explosing C functions in shared libraries. C's interface is very stable. C++'s varies from vendor to vendor and from compiler version to compiler version (GCC 2.95, 2.96, 3.0, 3.1, and 3.). A library in C plays well with Perl, Java, Python, C++, and many other languages. This is a good thing. (Mind you, this just means that the interface should be in C, but the implementation can be in C++. I certainly agree.)
Second C can be used dynamically. Not automatic dynamic linking, I'm talking about searching the filesystem for a library and using dlopen and dlsym (or LoadLibrary and GetProcAddress under Windows) to find functions you need. This technique is sometimes necessary and it's important that core libraries support it. C++'s interface is just too messy to support this.
MFC suffers from more fundamental problems than being a C++ wrapper over a C interface. MFC was originally developed fairly early in C++'s life and so had to work around weaknesses in the early language and in Microsoft's early implementation. Those original workarounds remain in place for compatibility, but significantly complicate the code. MFC tries to be all things to all people, and it tried to accomplish this goal from version one. As a result, it's a giant interconnected monstrousity with lots of non-obvious magic. For the longest time any use of MFC required you to use the Document/View framework, and that assumption still runs throughout the code. It's not possible to point to a subset of MFC and say, "this is the GUI interface as distinct from the application framework code." MFC is too bloody big.
Thin Windows API C++ wrappers exist and are much more pleasant to use. You can easily cook up your own in a few days and slowly expand it as you need it. If you want a more heavy duty C++ wrapper over the Windows C API, there are some of those as well. One of them is called Qt. :-)
Re:"Race KDE cannot win" (Score:5, Informative)
I agree with the basic thrust of the post, but in fairness this occurs with Gnome as well. Gnome seems less centralised to me, for instance they don't produce their own media player as such but RhythmBox is a Gnome app, they don't produce their own email app but there is Balsa and Evolution, etc.
There is way too much duplication, I agree. The projects are starting to work together a lot more now though, largely thanks to the work at freedesktop.org. Not just there though, for instance KDE was considering using GStreamer for its multimedia architecture at one point (I think they decided to wait for it to mature, which is fair enough).
Some KDE projects seem rather dead though, I think the more decentralised approach gnome takes (or rather, doesn't take) is a bit better. AbiWord isn't a Gnome app but you'd never know, it integrates nicely etc and is a good deal more active than KWord seems to be. Ditto for Gnumeric and KSpread. Noatun is just a joke, really, but it's kind of the "official" KDE media player.
Note that I have been KDE user in the past (alternating with less popular lightweight wm's), but Gnome seems to finally have gotten their stuff together with gnome2.
Agreed, at least in terms of desktop experience. It's not all there yet, but it shows great potential. KDE still leads in terms of developer platform though imho, their documentation is much better (though to be fair to gnome, they don't have a company like trolltech maintaining it for them). Also some Gnome technologies like Bonobo tend to be a bit confusing, especially in the more advanced usage. On the other hand, the KDE usability effort seems to be going nowhere quickly :(
Re:"Race KDE cannot win" (Score:4, Informative)
AbiWord isn't a Gnome app but you'd never know, it integrates nicely etc and is a good deal more active than KWord seems to be. Ditto for Gnumeric and KSpread. Noatun is just a joke, really, but it's kind of the "official" KDE media player.
You read in the article about there being something of a cultural difference between the two camps: USA vs. Europe, noisy vs. getting on with it. Well, this is the perfect example. You think AbiWord is far more active than KWord. It has more developers, more mailing list messages, more CVS commits, more releases. But look at the current in-development versions of both of them, and compare them with what they were like a year ago. I think you'll find the comparison doesn't come out in AbiWord's favour. Partly this is architectural - there's FAR more code sharing and reuse in KOffice/KDE/Qt than in AbiWord/GTK, partly because the balance of talking about it/doing it is further towards the doing it end with KWord than with AbiWord. I guess this also explains why Slashdot appears to have a tendency towards GNOME whilst the Linux community as a whole seems to prefer KDE.
And don't diss Noatun: you might not like it but from my point of view it's far nicer than anything else available. It plays all my music, has a good equalizer, does effects, the interface looks and works like everything else on my desktop (although it doesn't have to), and most important of all, happily hides itself down in the system tray when I want it out of the way and stays there. The KDE 3.1 version embeds Xine to play video: now it's the only media player I use. I love it.
On the other hand, the KDE usability effort seems to be going nowhere quickly
I take it you've not used KDE 3.1 yet then? There's some good improvements in there. And let's face it, GNOME usability still has a long long way to go *cough*GTK+ file dialog*cough*
Re:"Race KDE cannot win" (Score:3, Insightful)
Nope, can't agree with that. Firstly, I don't think Slashdot has a tendancy either way, I see just as many stupid trolls for and against both.
Secondly, for quite some time (dunno if this is still the case) KOffice had David Faure being paid to work on it. AbiWord hasn't had paid employees on it for a loooong time, so they're doing pretty well. Whenever I hear about people word processing on Linux, they're using OpenOffice or AbiWord. For spreadsheets, it's OpenOffice or Gnumeric. Maybe there is an utterly silent majority which actually uses KOffice, but I'd be pretty surprised. Mindshare does in fact matter, and I've found gnome/gtk based projects tend to have much better communication with the outside world than KDE based ones. AbiWord stable is getting pretty old now, but it's the one I'm interested in because I can see it developing and it's getting very cool very fast. KOffice on the other hand.... what is it doing? I do read dot.kde.org sometimes, but haven't seen a KOffice story for ages. The koffice development mailing list has about 200-300 emails so far this year, the same number as abiword development, which focuses only on a word processor, not a whole office suite.
FWIW I am European, and I think the characterisation by Gordon of the KDE developers as "get on and do it" and of others as "make lots of noise" is stupid - which desktop project was it that threw a hissy fit when Redhat altered some themes and changed some branding (something gnome also went through)? I know Europeans and Americans, and haven't noticed a great deal of difference between them in the way described.
I take it you've not used KDE 3.1 yet then? There's some good improvements in there. And let's face it, GNOME usability still has a long long way to go *cough*GTK+ file dialog*cough*
I take it you've not used GNOME2.1? I monitor what KDE and GNOME do, and so far GNOME has KDE beat by miles in the usability stakes. The GNOME/GTK world has a strong set of UI guidelines in the form of the HIG that even non-gnome apps are complying with (like xchat, gaim etc) because despite its lack of perfection, it gives a clear direction and it makes sense to follow it.
Last time I checked, I couldn't even find the KDE equivalents on usability.kde.org (though I have seen at least one before) and I've been told there are two such guides, both unfinished. The kde usability list seems to flap around without clear direction whereas you can visibly see and feel GTK apps get more usable.
And don't diss Noatun: you might not like it but from my point of view it's far nicer than anything else available.
You're certainly entitled to your view, but you're the first I've met who actually prefers it to XMMS. To be fair, the version in KDE3 is much improved over the one in 2.2 but it's still a hellhole usability wise. What made me laugh was the default plugin was called "Excellent" with the tagline, "A very simple, and therefore usable interface". First mistake, simple != usable. In fact, Excellent tripped me up several times, for instance it took me quite a while to figure out how to click on a part of the time slider to make it jump to a part of the song. Left clicking moves it in increments. Right clicking? No. Double clicking? No. Ohhh, that's right, middling clicking lets you move the slider to the part under the mouse. Obvious. Three types of playlist, none of which are compatible with each other and the only halfway decent one does NOT read .m3u files????!?
The only interface I found I could actually stand was the WinAmp one. Noatun does have some saving graces like the keygrabber plugin, and the notification tray icon is nice (though it doesn't actually notify you of anything). Nonetheless, I feel I'm justified in slamming it.
And let's face it, GNOME usability still has a long long way to go *cough*GTK+ file dialog*cough*
The file dialog is the exception that proves the rule, and a new one is definately going in for GTK2.4, that's guaranteed. It's harder for GTK because unlike KDE/Qt GTK is actually widely used outside of the desktop project. You can't just introduce random dependancies on Nautilus for instance. I'd also like to point out that the KDE file picker is simply a direct lift of the Windows one with all the UI faults it comes with, I won't write an UI review of it here because they exist elsewhere.
Really, GNOME has had to make some tough choices, ones that a small but vocal minority was opposed to every step of the way. Simplifying the UI was needed, GNOME 1.4 had an even more bloated UI than KDE in this respect, but KDE hasn't yet picked up the gauntlet. I don't think it is, but if Shawns theory holds true then it's simply because the KDE team don't like making controversial decisions, which is maybe why it looks and works so much like Windows.
Re:"Race KDE cannot win" (Score:2)
The reasons for this are largely consistent with what Shawn had to say about KDE: it is rather passive. I think this is a consequence of being extremely open - it's very easy to get involved with KDE, at almost any level because people will listen to you, but they will not stop and wait for you, but just keep going about their business of building stuff for KDE.
And to be fair, you have to give some credit to how difficult a truly open-source usability effort like this is. Usability is mostly about polish: consistency, intuitiveness, ease of learning for behavior, so it's best done by sitting down and going through the whole shebang at once, which is anathema to OSS and especially KDE's development process. The usability folks are now working to collect the usability reviews people have done into a database so developers can refer to them. With luck and some effort to get developers to interact more with the usability project, we can get more broadly involved.
If you want better usability in KDE, help us! [kde.org]
Re:"Race KDE cannot win" (Score:3, Informative)
Combine this with the fact that as of late, both GNOME and KDE are very intuitive and usable... Shawn may be right. When Red Hat 8 came out I tried the BlueCurve desktop and I really liked it. I started in KDE mode, but over the last couple of weeks I've been running it in GNOME mode and I really didn't notice a difference. That says something important.
Re:"Race KDE cannot win" (Score:2)
Trolltech's buisness comes from these people.
Re:"Race KDE cannot win" (Score:2)
Re:"Race KDE cannot win" (Score:2)
Are you kidding? If you read the LGPL anyone can take any "copy" of the software that is LGPLd and relicense it as GPL as they deem fit... there will still be an LGPL version out there, that you cannot prevent, but it may be used just the same as any GPL software as far as I care
KDE people also have the weird habit of producing their own versions of various pieces of software. Surely a conservative decisionmaker will choose a desktop-agnostic Mozilla or OpenOffice over the KDE-specific versions. KDE applications might do better by just dropping the K from their names, thus competing on their own terms (snappines and other virtues associated with Qt).
Well Mozilla is really huge... Konqueror is pretty tiny in comparison. Konqueror also has the ability to use all the good KDE technology like KIOslaves [kind of like libferris in functionality] transparently... KDE is not just a desktop its a complete framework with which to develop applications. A fairly rich one at that. Why would anyone running KDE not want all KDE apps in this light?
Re:"Race KDE cannot win" (Score:2)
You're right KDE can't win, if you're going to keep shifting the goal posts like that. As long as KDE was not GPL, all you heard was 'Unclean! Unclean!' . Now that KDE is GPL, that isn't good enough either? For heaven's sake make your mind up.
As for preferring KHTML over Mozilla components, I think Apple pointed out last week [kde.org] exactly what a good decision that was. Gecko is a good component, fair enough - but KHTML is at least as good and very much smaller.
Re:"Race KDE cannot win" (Score:2)
Except...mozilla's UI is butt-ugly and totally wrong for ANY desktop. The default UI themes are WRONG for both Gnome AND KDE and neither desktop system's theming/coloration system works with mozilla. You have several apps up and running, they are likely to all look compatible with each other instead of a random scattering of UIs and colors. Start up mozilla and it looks either VERY much like a windoze app or some blue thing with no compatibility with your system look/feel/theme.
Openoffice/staroffice...same thing only you get one and only one UI - a windoze interface.
I select my desktop colors and themes for a reason. They work best with ME and the colors work best with me. It looks coherent and is the same across apps (I use KDE). I start up a gtk app or mozilla or staroffice/oo and they are bastard stepchildren clashing with everything. Same goes in Gnome, which I have run on occassion when I've trashed KDE during an upgrade or rebuild or some such. Those two apps don't look right in Gnome either.
Publish your API and let the desktop handle the GUI part. I want the apps I use to obey my color and theme selection, period. I made the selection for a reason. Lowest common demoninator GUI design is butt-ugly.
Re:"Race KDE cannot win" (Score:2)
In KDE 3.1, change kcontrol->desktop->panels->menus->Menu Item Format to "Description (Name)".
whee..
Interesting company concept (Score:5, Interesting)
His employees were all basically referred and the traditional face-to-face interviews were obviously never done.
It's a new way of doing business. I like it.
Re:Interesting company concept (Score:5, Funny)
...
It's a new way of doing business. I like it.
Nonsense! hitmen, crack dealers and major crimelords have worked this way for years!
- Oisin
Re:Interesting company concept (Score:5, Interesting)
His employees were all basically referred and the traditional face-to-face interviews were obviously never done.
It's a new way of doing business. I like it.
Exactly what I thought when I read the article. I have often toyed with the idea of employing someone over the net, but I've never had the balls to actually do it. It seems to me it must be a very efficient way to run a business.
I'd be interested on what kind of contracts he uses. Does he employ these people full-time, or by project? Another issue - getting stuff delivered on time. I guess if you make products like theKompany.com, it doesn't matter if you slip. But working for clients, you have to deliver when you promised to, or you'll quickly go out of business. Is it practical to run a business this way if you need to deliver to clients with tight deadlines?
Re:Interesting company concept (Score:2)
I too have considered hiring people over the Internet to telecommute but I would want some form of personal assurance that they'd get the work done to the best of their abilities and not break any laws in the process.
And yes, in hiring programmers, I'm concerned that someone who is never physically met or supervised might not be on the up-and-up, although I'm pretty convinced that over 80% would be.
Re:Interesting company concept (Score:2)
that said, what makes having an employee take up a 10x10 office space (costing you ~10k year), your face to face interview and all that assure that stuff is going to get done to the best of their abilities, on time and in a legal manner? i'd say give it a shot on a temp basis each time and if it works out, go longer term.
Re:Interesting company concept (Score:2)
Furthermore, aren't there some great tax-avoidance (not tax evasion - that's illegal) measures you could take with such a company? It sounds like an idea case for an off-shore company, as you're not employing anyone in your country of residence and the company doesn't need any physical premisies.
He's lucky [was Re:Interesting company concept] (Score:4, Interesting)
It was a nightmare, communication was horrible. Sure a lot of problems boiled down to mis-management but it was certainly compounded by the fact that we were far apart, and people could basically do whatever they wanted and get paid for it. There was tons of turn over, they kept hiring incompetent people and firing them. Almost sort of an expensive trial and error.
Re:He's lucky [was Re:Interesting company concept] (Score:2)
I guess that's why he uses personal recommendation only.
CodeWeavers and Ximian do something similar, though not quite to the same extent.
Re:Interesting company concept (Score:2)
Write Once, Run Anywhere? (Score:2, Interesting)
Sound familiar? That's why Java is so popular. The very problems described in this interview are the kind of problems Java and its standard class libraries resolve.
Re:Write Once, Run Anywhere? (Score:2)
java gui's are still quirky. yes they are portable. i like the fact that netbeans looks the same on linux as it does on windows. i like netbeans itself as a development tool, but it uses resources. if you can't get your manager to spend the 100$ to upgrade your ram because of the red tape that goes with a memory upgrade in an organization, you go with eclipse which uses native widgets.
Re:Write Once, Run Anywhere? (Score:2)
When is the Linux community going to wake up to what a pitiful mess software distribution is? Linux will never be a popular desktop OS until this problem is solved. How are you going to explain to a naive end-user what it means when they can't run an app because it requires foolib 2.0, and they have foolib 3.0 (Mandrake version) installed?
Java is one solution. But in truth, the shared library disaster is specifically a C++ problem. The problem is that unless you're extremely careful, any modification you make to an exposed class in C++ is likely to cause a binary incompatibility. This isn't a problem with Perl, for instance (although your Perl libraries might have C++ linked in to them, in which case you still have the problem,).
And that's assuming you're trying to avoid binary incompatibilities. What's really sick is people like Red Hat intentionally introducing these incompatibilities. It starts to sound a lot like the old world of proprietary Unix, where you'd be locked into one vendor.
Re:Write Once, Run Anywhere? (Score:2, Funny)
Re:Write Once, Run Anywhere? (Score:2)
Shh... don't tell the MoneyDance [moneydance.com] people!
Can't we all just get along (Score:5, Insightful)
Here's the real problem:
"... focus on Gnome and the relegation of KDE 'to second best', other Gnome vs KDE issues
With Mandrake focusing its attention on finances [slashdot.org] -- it is time for a leader such as RedHat to do what my father used to say to my brothers and I when we'd be squabbling over this-n-that "I don't care who's fault it is, I'll knock both your heads together -- now play nice!"
So long as we have these pissing battles between Gnome and KDE -- Windows will continue to enjoy its top of the heap status.
Re:Can't we all just get along? (Score:2)
"We know that Red Hat has never been a fan of KDE, nor has it ever supported KDE. As I recall, they financed the original work on Gnome and despite what a large company they are, there was only one KDE advocate at the company who quit last summer and has since made a forked version of Red Hat called Ark Linux if I recall.
Lets face it, the operating systems are going to drive the interface, and the big companies going to drive the O/S -- how else could a buggy system such as Win95 have been inflicted on us all?
I admire the work gone into KDE, but perhaps its time to make an alliance that will save everyone concerned on the Linux side?
Re:Can't we all just get along? (Score:2)
i'm not quite sure, but it certainly didn't have anything to do with Microsoft forcing vendors into exclusive contracts to install their software on the hardware they sold. big companies didn't drive the O/S, IIRC A big company pretty much forced it there.
Re:Can't we all just get along (Score:2)
Re:Can't we all just get along (Score:3, Informative)
Competition is what we need (Score:5, Insightful)
But the most important thing is that the competitive enviroment is maintained. If one get's to dominate too much, there's no real need to really invent stuff. Just look at what happened to Windows I haven't really noticed much of a difference since NT 4...
Re:Competition is what we need (Score:2)
But supporting your argument, what caused that step? Mac OS X, otherwise known as the competition. Indeed, competition is very good for the desktops.
Support QT..... (Score:5, Interesting)
With QT it's either open up you code under GPL, or pay us. Which I think is by far the best option.
Also QT runs on hand helds and the like and it's C++
Gnome winning? (Score:2)
On another thought, perhaps we need to change the attitude we have towards making applications. Many have considered it, and some do it - separating the program from the interface, Qt or GTK. What if we create the program like games are often made today? An engine with an interface, but a powerful sdk for developing your own mods and changes. So if we have a powerful e-mail client such as evolution, have all the code and features done including interface, but designed in such a way that any new programmer could come along and create a Qt, OpenGL or whatever interface himself. We separate the code from the interface, with functions to hook them together.
This could potentially tie in with plugins, making them easier to create, since most parts of the program would be made easily accessible to the interface for new programmers.
Re:Gnome winning? (Score:2)
It seems complex, until its hidden in the libraries and one can ignore one's GUI code in one's app code completely.
Re:Gnome winning? (Score:2)
I dunno. I've not seen any compelling stats for which is more popular. Anyway, it's odd but when I uesd KDE I used mainly KDE apps, and now I use gnome I mainly use gnome apps. Except I did use some GTK/GNOME apps in KDE, and now I don't use and KDE/Qt apps at all, most of the time when I find an app I want it uses GTK. Dunno why.
On another thought, perhaps we need to change the attitude we have towards making applications. Many have considered it, and some do it - separating the program from the interface, Qt or GTK.
A lot of apps do this, there are lots of front ends to LAME, or cdrecord etc. But yeah, it's something to look in to.
If Only... (Score:2)
Back to topic, KDE is a top rate set and Qt is a top rate library. Its nice having a choice for how I want to manage my desktop. This choice may beat you the death with all the different builds you have to do - so be it.
J.
Keep them both (Score:5, Insightful)
The problem with Windows has been that there was no real competition. That problem is being solved. I know that there are folks out there who are devoted to Debian and hate what Red Hat has done with 8.0, but regardless, I could hand the Red Hat discs to any of my family members and they could install it on their computer without wiping out the Windows install. This is one thing the Linux community has been shooting for. Are there problems with RH 8.0 and BlueCurve? Sure, but it's something that compares well with XP.
I like having KDE and Gnome to choose from and, at this point, I don't know enough to choose one over the other forever and ever amen. Right now, though, I have the choice and that's what brought me to the party in the first place.
Good point, and a question (Score:2)
I recently downloaded Gentoo (NICE!) and in reading the emerge docs, it states that you can set a flag in the source compilation process to include KDE and Gnome functionality into the build scripts for apps that support them.
This leads me to believe that I can run a Gnome app in KDE and still have all the Gnome-y goodness within that application. Ditto for running KDE apps in Gnome.
Is this true? If so, why worry about using one desktop or the other exclusively? Is this something the only Gentoo does, or does it just handle it more gracefully? What are the disadvantages of mixing like this?
Just a little confused. Thanks!
GMFTatsujin
There's a lot of misinformation about KDE. (Score:3, Interesting)
With regard to Red Hat and Bluecurve... Those of use who use KDE do not think that everything was well in Red Hat 8. I'm a longtime Red Hat user. I installed Red Hat 8 and tried to make their KDE work. I gave up. I ended up going to ftp.kde.org to see if there was an "official" distribution of KDE and Qt for Red Hat 8 that would repair the obvious rendering bugs, have working fonts, either function with standard KDE icon themes or use a non-broken set of Bluecurve icons, and would actually use KDE applications for app-to-app functionality instead of GNOME ones.
But no dice, apparently the official KDE packages for Red Hat 8 are made by Red Hat, so the KDE I downloaded was essentially the same (and essentially as broken) as the KDE on the Red Hat 8 CDs. I switched back to Red Hat 7.
And what's with all the license complaining still? Qt is dual-licensed, GPL or commercial with paid development. This should be enough to keep the GPL zealots happy and the "GPL sucks for companies" people happy... but apparently some people will not rest until an inferior product, GNOME, has been declared the winner by the 800-lb gorillas of the Linux world and KDE has been marginalized.
The fact that KDE is still as good as it is and is still as popular as it is demonstrates just what a great product it can be (notice I said "can be" because we all now know just how much depends on distributors).
Re:There's a lot of misinformation about KDE. (Score:2)
So.... what is your point again? If you prefer a KDE centric distro, use SuSE or Mandrake.
GNOME2 feels unfinished because, ... (Score:3, Informative)
SuSE has not broken these packages with their changes, they were crap to begin with. Redhat broke KDE with their changes.
I know that OTOH KDE releases feel finished, because I compile them myself and there is nothing broken or unpolished in them. KDE-3.1 will be better than ever!
My favourite.. (Score:2)
GNOME has hits flaws and quirks, but I just get up and going on GNOME, and can customize it to my needs a hell of a lot faster than I can with KDE.
Re:My favourite.. (Score:2)
KDE has hits flaws and quirks, but I just get up and going on KDE, and can customize it to my needs a hell of a lot faster than I can with GNOME.
Sorry, I just had to do that =)
Re:My favourite.. (Score:3, Insightful)
What an odd thing to say. Most people who have been using Gnome and KDE since they started would say only now has Gnome about caught up with KDE in quality. After all KDE had a huge head start. Also ugly? In what way? They both look pretty much the same to me. Either desktop can also be made to look like anything from Windows XP to OS X to Next etc etc.
The only people i have ever heard say one is "easier" to use then the other are the completely biased ones. They are both easy to use and have been for quite some time.
Your post reminds me of the early days when all the KDE bashers used to say "but KDE isn't themeable like GNOME!".
"GNOME has hits flaws and quirks, but I just get up and going on GNOME, and can customize it to my needs a hell of a lot faster than I can with KDE."
Again good for you, but customize it? Ever hear of KDE's control panel? It doesn't get any easier than that. It just sounds like you made up your mind a long time ago that Gnome was somehow "better" than KDE and your either unwilling to learn more about KDE or are just hopeless biased.
Today they both about equal and anyone newbie could get their work done easily with either desktop.
Gnome will always be favored by corporations (Score:2)
I am a researcher at a large university. We develop software for numerous entities including national labs, industry, and acadamia. About a year ago, we wanted to start a new tool with a better user interface (we've been using Tcl/Tk for other projects). We started developing with QT but quickly had to abandon after our supervisor pointed out the GPL issue. You see, we can't GPL our code even though we nearly always open-source it under our own license. QT libraries our either GPL or you buy a license. Our funding sources prevent the use of GPL libraries because we often sign agreements with them such that they can use our code in commercial products. We have the money to purchase QT licenses, but dealing with the budgets (bureaucracy) and licences with other organizations is just to complex and unwanted by most involved (except for some developers).
This is the exact same reason why Gnome has more commercial support (at least in the U.S.) than KDE. Sad but true.
I can't stand KDE or GNOME (Score:2)
That, and the fact that neither GNOME or KDE is worth the time of day as far as I'm concerned. They're both bloated Windows-wannabes. I'll pass. No thanks. I gave at the office.
Anyone out there who wants to try something that's ACTUALLY different, check out Enlightenment [enlightnement.org] (which is what I use for Linux) or AfterStep [afterstep.org] or WindowMaker [windowmaker.org]. Real people providing real alternatives to the twerps in the GNOME and KDE battalions. Psssst, you can even use GNOME and KDE software in them...
Re:I can't stand KDE or GNOME (Score:2)
> Both camps have a lot of vocal jerks in them, who apparently hate each other for NO GOOD REASON.
Well, the fact that GNOME was mainly started as a reaction to KDE probably has something to do with that.
Re:How does commercial Free Software work? (Score:2)
Re:Little of about Debian (Score:2, Informative)
Re:Little of about Debian (Score:5, Funny)
So you're saying I should try Debian again sometime around 2005?
Re:Little of about Debian (Score:2, Informative)
Supporting a stable version of Debian requires supporting Qt 2.x, KDE 2.2.2 and gcc 2.9.5 whereas supporting the various other major distributions is a matter of supporting Qt 3.x and KDE 3.x and gcc 3.2. There must be a separate release for Debian stable or make the user install extra libraries.
Supporting Debian unstable as a platform would be ridiculous, even if the gcc 3.2 transition and kde 3 upload had already occured. Unstable usually breaks several times before release.
Debian is and always will be a bit behind the rest in terms of up-to-dateness.
Re:I hope it goes well in the future. (Score:2, Interesting)
Re:KDE vs. Gnome (Score:3, Informative)
The default desktop in Red Hat Linux has been Gnome for a while now - and never KDE. They have both been included since Qt had a GPL option - before that, the licensing issue made that a no-go for serious vendors.
Re:Best KDE Distro? (Score:2)
But it really depends on what you want, if you want a "only the good sides of Windows"-distribution Mandrake's the best joice, if you want a "corporate-user"-distribution you should take a look at SuSE, if you want a "just as cool as debian but with a current version KDE without hunting for inofficial sources"-distribution I recommend gentoo (which is not biased towards KDE but isn't against it either (as Redhat arguably is)
jm2
Re:Best KDE Distro? (Score:2)
Redhat is also fine, as long as you install third party KDE rpms or build it from source (and don't use the supplied version of KDE)