Following Novell’s announcement that they will be combining the best of KDE and GNOME, Heise On-Line is now
reporting (Google
Fish) that Chris Stone let it slip during his keynote at BrainShare 2004 that Novell has chosen to standardize on Qt as development environment. If the latest SUSE desktop is anything to go by, we can expect an integrated desktop based on KDE & GNOME out of this.
More on Novell’s KDE/GNOME Desktop
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Please, Novell, buy off Trolltech – use their engineers to merge Mono into Qt (or the other way around), but nontheless, make Qt free (the act of buying Trolltech means that Qt would automatically be given out under the BSD license free). It would better itself in competition with other toolkits, APIs.
But SUSE its already KDE based, so qt based, and its tools are already qt based?
What do they mean with that?
qt based GIMP?
How would QT automagically be under the BSd license if it’s GPL’d ? already and why would they BSD it if that means that others can grab the work and make money off it without their consent ? If anything Novell would GPL or LGPL it before using a BSD license for QT.
Great. Qt as the default toolkit… well, let’s all KISS OUR USABILITY GOODBYE, as Qt/KDE’s culture of crack-tastic, feature-bloated, “geeks only” user interfaces invade the Gnome software we all know and love
Perhaps I’m over-reacting, but this really seems to spell doom for the whole concept of Linux having a *usability* edge over Windows, not just a technological one. Don’t get me wrong, I’m very glad KDE exists, since no one interface is right for everyone, but it’s /not/ suitable for most end-user types. It’s also not suitable for people like myself who, while very competant with computers, would rather not have to use a bloody /search engine/ to find the right preference for a given program on the desktop (ie: the Control Center).
This also calls into question, “Why the f*ck did they buy Ximian in the first place” if they plan to rework all their stuff with Qt?!? Why not just buy SUSE by itself?
Maybe it’s time I give up and buy a Mac… *sigh*
> This also calls into question, “Why the f*ck did they buy
> Ximian in the first place” if they plan to rework all their
> stuff with Qt?!? Why not just buy SUSE by itself?
I think that Novell didn’t really knew WHAT exactly they bought here. My guess is that they initially thought that they bought a Desktop (e.g. they bought GNOME) or that they bought MONO. Or they bought the experts knowledge of those working on MONO. I can’t exactly tell it but guess it only.
@Anon
“Perhaps I’m over-reacting, but this really seems to spell doom for the whole concept of Linux having a *usability* edge over Windows, not just a technological one. Don’t get me wrong, I’m very glad KDE exists, since no one interface is right for everyone, but it’s /not/ suitable for most end-user types. It’s also not suitable for people like myself who, while very competant with computers, would rather not have to use a bloody /search engine/ to find the right preference for a given program on the desktop (ie: the Control Center).”
Yes you are indeed overreacting here. The situation with GConf on GNOME isn’t better either (really) as soon as it comes for specific preferences people show up in the irc channels asking how they can change things. And then it’s time even for us to search through hundrets of GConf keys before we find something we can offer the user as solution. Often we know by guessing where we can find the key that helps the user but sometimes we on our own keep searching with GConf editor.
You also shouldn’t forget that KDE as is in CVS right now has heavily cleaned up Menu entries, Toolbar and well organized the Settings. At the end you only go through Control-Center one time and voila it’s easier than having to use GConf to change keys and then all these standalone capplet modules as GNOME does it now. I always prefered the Control-Center way over the standalone capplets (and extra GConf modification) way and it looks like the idea with the Control-Center is being adopted as very good solution on different architectures as well. See MorphOS as good example here.
That can’t be seen as good news for GNOME. For a while there it looked as though the standard might become GNOME, but that seems to be slipping away more and more now. It’s a shame, KDE is just plain ugly as sin. Of course if all the apps were written using qt then it wouldn’t be such an issue, but so many GOOD solid apps like mozilla, gimp, etc are gtk2 ones. I also see gnome as being a much more open project than KDE. But, whatever.
I’ve heard that some KDE developers once started to port the GIMP to QT (called KIMP) but the GIMP team was strongly opposed so they cancelled the project.
That’s stupid really. It’s open source, correct? Why stop them from doing it? It’s a banner GNOME app and must stay that way? Silly.
This also calls into question, “Why the f*ck did they buy Ximian in the first place” if they plan to rework all their stuff with Qt?!? Why not just buy SUSE by itself?
Agreed. It doesn’t make _any_ sense (I’d like someone to explain how it would).
Then again, I haven’t read the article, because I can’t find an English version (Heh, it’s all lies I tell ya!).
Great. Qt as the default toolkit… well, let’s all KISS OUR USABILITY GOODBYE, as Qt/KDE’s culture of crack-tastic, feature-bloated, “geeks only” user interfaces invade the Gnome software we all know and love
I prefer GNOME, and I think KDE is whack. But when it comes down to it, I just want to see consistancy and an end to the multiple API bullshit. It’s stifling progress. The Linux desktop would have been in full swing 4 or 5 years ago it wasn’t for this. Sometimes “choice” is not a good thing. Mono would have helped solved the problem of
“choice” a bit, so choosing QT doesn’t make any sense to me.
Don’t get me wrong, I’m very glad KDE exists, since no one interface is right for everyone, but it’s /not/ suitable for most end-user types.
That’s not necessarily true. One of the things about KDE is that it is extremely easy to customize. A little bit of elbow grease on Novell’s part could result in a very polished KDE desktop. Unlike with GNOME, with KDE its largely a matter of editing a bunch of XML files or Qt Designer files.
This also calls into question, “Why the f*ck did they buy Ximian in the first place” if they plan to rework all their stuff with Qt?!? Why not just buy SUSE by itself
Mono, Ximian Enterprise Connector, Evolution, Red Carpet. Also, Ximian had many people with desktop Linux experience that were valuable to Novell, as evidenced by the fact that 20 Ximian guys were transferred to SuSE to work on the Novell desktop. For what they paid, Novell got their money’s worth. Ximian cost $20m, while SuSE cost $210m. It is expected that SuSE would be more central to their strategy than Ximian.
Using Qt is a very solid choice. It looks like Novell is really going to give Microsoft a run for their money.
Now all Novell has to do is get the bravos from Ximian back into their cages and coding instead of blogging.
I don’t know the details but I’m pretty sure it has nothing to do with GNOME. The GIMP is far from being a GNOME app (just take a look at their interface). However, GTK stands for “The GIMP ToolKit”…
Then again, perhaps the mainteners just abandonned their port as the GIMP is coded in plain C.
Its hard to say that GNOME is a more open project than KDE. GNOME, in 2.x, has become a rather top-down project. Now, this has certain advantages. For example, the HIG-ification of GNOME could never happened without this sort of approach. On the other hand, it also caused a lot of users to migrate away from GNOME in frustration. As a result, the KDE community often seems a lot more vibrant and involved with the project than the GNOME community.
> I also see gnome as being a much more open project than KDE.
I unfortunately can not confirm this. I participate for around 5 years to GNOME now (including own programs, fixes, patches etc.). While GNOME does use an open source licensing model the project itself is anything else than true open. As long as people won’t listen to users feedback, as long as people have their own little regime there, as long as slandering others who have a different opinion is a matter of daily business as long I can’t call it “open”. Open is something where people learned to get along with, where others are respected for what they contribute, where they listen to other people. Where you as individual have the same rights than others.
GNOME is not just the Desktop. It’s the entire community that means. The Desktop, the people, the fun. If something of these things don’t match then it’s probably not open anymore. Unfortunately there are a handful of people full of rudeness who took the right to play god’s in their own hand.
When I compare the KDE and GNOME community then I must tell you that I was never ever attacked that way as person, as human by the KDE people as I was from certain GNOME people. In my 23 years of working with computers and professional development I was never treated that ugly than on GNOME. The other communities that I participate in (Amiga, MorphOS, QNX, KDE, Console stuff) I am treated as a fair person, the community is nice and fine. The people are helpful and altogether fine persons to talk with.
Sorry for refering to my person all the time but I wanted to give you a fair answer of how I saw the things on my own for the past years. I also know other people who can say similar things about the GNOME project (at least they told me, you probably can’t count them to make everything public like I did).
I believe that Novell made a good decision with using KDE as their architecture. They simply went for a good framework and where they see hope of rapid applications development, good solid stuff etc.
# Qt as the default toolkit… well, let’s all
# KISS OUR USABILITY GOODBYE,
You obviously have no clue what you are talking about. Qt in fact encourages good usability by encouraging the programmer through their tools and API to create a consistant and usable interface.
I don’t see how gtk which is technical far inferior (and for which development has slowed down now that Redhat gave up on the desktop) would do the same.
# whole concept of Linux having a *usability* edge
Sorry, but the claim that Gnome is so much more usable is outright laughable considering that they have been promising a usable file selector for years. But Gnome has a track record for breaking promises and just following the hype of the day: While Gnome has started as an effort to create a “gnu object network model environment” there is almost nothing left of the so much hyped promises about remote (corba) components everywhere. Today they promise Mono (which definately will lead us into another SCO-like patent desaster) and tomorrow they will hype something else.
They aren’t even able to make sure that the applications share their addressbooks and bookmarks properly. Gnome as it exists today is just a collection of applications which share the same look. And there are a lot more issues I don’t even want to get started into.
# edge over Windows, not just a technological one.
The problem is that with gtk we would only be far behind Windows in terms of technology
# but it’s /not/ suitable for most end-user types.
Says who? Then why do most existing users like KDE? Why does KDE always win nearly every single desktop poll? Why is it that Linux is most successful on the desktop in areas and actual deployments which are using KDE (Europe, Munich, …)? Why do most distributions use KDE as the default? Why did they buy SUSE which has always been a desktop distribution and which has built their success on KDE?
# This also calls into question, “Why the f*ck
# did they buy Ximian in the first place” if they
Because at that time they didn’t know that later they would be able to buy SUSE and therefore part of the #1 desktop KDE as well.
Personally I don’t welcome the decision to focus on one desktop because I like competition and choice. But if this really has to happen KDE is without doubt the better choice for users as well as developers.
IBM & HP have chosen Novell SUSE Linux as their Linux distro of choice. Reason (HP spokesman): Novell is a long-time business partner and Red Hat doesn’t offer a desktop Linux.
Novell spokesman: Customers tell us they want one desktop enviroment. So we will migrate to one desktop.
Novell has decided to use QT for their desktop developments.
Somehow they want to combine both desktops into one. I don’t see how this should work, though.
GNOME has nothing to do with Mono, thats Novell/Ximian.
No need to panic, but if true, this is not good news for GTK+. The fact of the matter is that this is a brilliant move on SUSE’s part because Qt is C++ by nature and it has a dual license for those wanting to release propietary software on a free platform such as SUSE Linux. So developers who are used to programming GUI applications in C++ (the MFC marines) can come to SUSE and expect that they can continue to use their valuable experience on a C++ toolkit that is rumored to be the best of the best in the industry. This would also quell the recent push from Pennington to standardize on one high level GUI toolkit (of which Qt was not mentioned, IIRC). Granted C++ is not a managed language, but it does have a standard object oriented syntax, unlike the pseudo OO syntax you will find scattered around the GNOME C libraries. Not to mention Qt is quite portable across Windows/Unix/Linux. Also, C++ is fast and nobody complains more than when a desktop GUI is slow (which is frequently the case with badly programmed Swing apps). I’m sure there will be some sort of political uproar as a result of this decision, but personally, I think it is the right one to make from a business standpoint. This doesn’t mean that Gnome will die (or does it?). It would be a shame if Ximian up and ported everything to Qt, but I don’t see this happening in the near future. Regardless of what SUSE does, the Linux community will always have choice and that is the beauty of free software. But, it would also be a shame to see this diminish interest in Gnome, which is really coming into its own lately, despite its confusing programming interfaces. The next couple years should be interesting. It will be interesting to see how Red Hat responds to this. At the moment Red Hat still has a lot of influence, but Novell has made some brilliant decisions as of late and seems to be showing some serious leadership. As an aspiring ISV, SUSE is increasingly becoming THE platform to target if one is trying to find a stable platform (and a stable partner in the form of Novell) to deploy one’s software on. It seems as if the Linux world is splitting into two distinct and unique factions: the corporations and the do-it-your-selfers. Which may or may not be a bad thing depending on how well compatibility is maintained among both groups. Keep it together y’all.
Novell buys Trolltech, LGPLs the thing, gets the QT-mono bindings up and running and then sweetness.
So, if this is really true…I had a hard time understanding the google fish…I wonder how Ximian plays into this. Seriously, what is the point of Ximian if they’ve standardized on QT. Evolution? doubtful. Mono? Possibly, but they pump some resources into getting the QT bindings for Mono then.
If this is true, then I’m a bit surprised. QT is a great toolkit and all, but I’m just surprised that they would be so dependent on trolltech for their desktop strategy. Novell has no say in what happens with the QT toolkit, with Gtk+ at least they could enhance it to suit their needs.
By the way, any germans want to do a real translation of this thing?
It’s a shame, KDE is just plain ugly as sin. Of course if all the apps were written using qt then it wouldn’t be such an issue, but so many GOOD solid apps like mozilla, gimp, etc are gtk2 ones.
I don’t understand this either. There are far more developers for KDE: why can’t they make their applications look good? They should have the time and the ressources… or maybe they just keep wasting their time proselytizing GNOME developers like one of our friends here?
I don’t use GNOME because it’s the ultimate desktop. I use because most – if not all – GTK/GNOME applications are superior to their KDE counterparts, with the notable exceptions of K3B, KDevelop and Quanta. I really don’t understand why as KDE should have the advantage. Then again, I’m pretty sure some KDE users think the opposite so it can only be a question of personal preferences.
“The next couple years should be interesting. It will be interesting to see how Red Hat responds to this. At the moment Red Hat still has a lot of influence, but Novell has made some brilliant decisions as of late and seems to be showing some serious leadership.”
The next couple of years from my personal view. GNOME will move on the same way it did the past years. Fixing stuff, trashing half finished apps for new approaches, getting new technology into the GNOME plattform which then stays unused or get half implemented (HALF!!!). They will keep developers busy catching up to their permanent changes of deprecating functions and adding new functions.
What influence should Red Hat have ? They don’t own Novell nor do they own SuSE or IBM. Even Red Hat has given up on their Desktop solution by splitting it out to an own project called Fedora which is maintained by semi-/nonprofessionals.
– Red Hat given up on the Desktop (split to Fedora),
– HP jumped off of GNOME,
– SUN working on their own GLASS desktop (not GNOME based),
– Eazel died long ago and now working for Apple and KDE components (Safari and KHTML),
– IBM, SuSE, Novell and many other named companies using KDE these days.
# GNOME has nothing to do with Mono, thats Novell/Ximian.
That’s just wishful thinking on your side. Name one single technology which has been hyped by Miguel Icaza which did _not_ go into the Gnome core in the past. The usual trick here is to find more mindshare for his technology to make its inclusion inevitable and to gain more people in favour of it than he would loose …
But feel free to cheat yourself with your belief that Mono will never go into Mono.
> It’s a shame, KDE is just plain ugly as sin.
Depends. There are good Themes for KDE as well as good Themes for GNOME. But I believe that the Plastik KDE Theme is amazing. It’s one of the best Theme engines I’ve seen for the past years (besides High Performance Liquid and Industrial which I like too).
http://www.kde-look.org/content/preview.php?preview=1&file=7559-1.p…
http://www.kde-look.org/content/preview.php?preview=2&file=7559-2.p…
My apologizes this is an older version of Plastik Theme for KDE. It comes quite close to Eugenias Theme mockup she made half a year ago or something. The plain simplicity and aesthetics. The Plastik Theme found in kdeartwork (which is yet again a heavily improved version) is looking even far better for KDE 3.2.x you should test it.
Amazing analysis, Ali, you’re really showing off your detailed knowledge of the GNOME project. Let’s tackle some of these, just for the edification of some of your readers:
* Red Hat axed their “Red Hat Linux” product, which has no correlation with their desktop efforts. They have now committed to a number of versions of their “Red Hat Enterprise Linux” product, including RHEL WS, which is 100% a desktop product. Indeed, Red Hat recently hired a surprising number of hackers for their Desktop division.
* HP certainly gave up their port of GNOME for HP-UX, as they understood that the *nix desktop market would be dominated by Linux. Great move. They’re now shipping various distros, and to my knowledge, using GNOME as their default desktop. They are keenly involved in the GNOME Foundation Advisory Board.
* Looking Glass is a really awesome experimental project researching potential not-quite-desktop metaphors that could be used in the future. It is Java based, and is essentially a 3D window manager. Definitely something to look out for, but you won’t see Sun changing their Java Desktop System strategy just yet. 🙂
* Eazel, unfortunately, never had a strong business plan. It is certainly true that many of the ex-Eazel hackers are now on the Safari team, and chose KJS/KHTML for it. Top stuff, it’s a great product. 🙂
* Lots of people shipping KDE, lots of people shipping GNOME, even more shipping both. We have a really vibrant desktop effort going on here, and two strong projects for the community to get involved with.
I tried 3.2.0 about a month ago and I wasn’t that impressed. It’s hundreds of times better than Keramik, though. I’d rather use twm than KDE with Keramik…
Personally, I think the KDE/QT widgets just don’t look as good as the ones with GTK. Furthermore, Pango seems to render better fonts than KDE (although both are supposed to use FreeType… I don’t understand). However, my main grudge against KDE is the quality of the applications. Not necessarily the quality of the code, but the quality of the interface, etc. Like I said in a previous thread, users don’t really care of the framework. They just want something that works.
I’m no KDE basher. In fact, I’ve used KDE for quite a while (from 1.3 to 3.1) and I’ll probably switch to KDE when I’ll begin application development on Linux… I just don’t understand why they don’t care of the look and why they merge a boatload of crappy applications with it. Hopefully, Novell will start collaborating with TrollTech and the KDE team to solve these issues.
Woot woot go quality team!
http://quality.kde.org/
“Novell buys Trolltech, LGPLs the thing, gets the QT-mono bindings up and running and then sweetness.”
Exactly what i’ve been thinking. Except maybe IBM buys Trolltech (i bet there have been negotiations on this already), LGPL’s all 3 Linux/Mac/Windows QT platforms, open sources the QT Designer (with RAD tools). Very astute investment for IBM. QT would be the defacto Linux toolkit.
Ximian’s role would be mainly Mono, which is extremely important for Novell and Linux. How many books are there on QT, or GTK – a handful? Now how many are there on C# and .NET – tons.
Mono “freeloads” on all this .NET documentation, and windows developers get at least this: the ability to transplant their ASP.NET web apps straight to SUSE Linux, no code rewrites; they use a language they know already, C#; and they get Mono-QT Bindings, so there’s no need to drop into C++.
Finally, you get those Mono-QT bindings ported to Mac – wow!
I can’t disagree with anything you said. If GNOME is to ever become more than a hobbie project, some of the developers need to eat some humble pie. No doubt these are brilliant people, but they need to realize that the perception of GNOME as an always broken platform needs to change, how fast this happens is anyone’s guess. No amount of marketing will change the perception, that means GNOME developers need to lower barriers of entry into the GNOME API Skull & Bones society. I believe some of the leaders of the community are trying very hard to simplify things, and I believe they will succeed in the not too distant future. It seems Novell has gained tremendous influence over the community by the simple fact that they now employ some very key members of the Linux community. Thus, it seems they now have the power and influence to say this is how they want the community to focus itself starting mañana. Is this a trend we are going to see increase? Free software developers being bought to do things which are in the corporations best interest? It seems there’s a fine line between scratching an itch and scratching the CEO’s itch. This is a very interesting story indeed. Perhaps some projects such as GNOME shouldn’t aim for corporate acceptance, maybe they shoud just try to ‘keep it real’ and focus on the needs of the free software community, whatever that means now. It’s becoming increasingly foggy now as to what free software means as more and more free software developers are employed by rich and powerful corporations. Politics are politics, the code may remain open but that doesn’t mean it will remain free. The GPL can defend the code, but it can’t hold of the wheelers and dealers and those tempted by fortune and fame. The ring is powerful indeed.
I don’t think GNOME or KDE are going to go anywhere soon.
To me, it sounded more like “Novell is going to write their own apps (think Netware) in Qt” rather than “we will search and destroy GTK+”. This seems like a fair decision if they’ve got a surplus of Qt hackers from buying SuSE. They’re still pressing full steam on XD3, which should give some indication that they’re still quite interested in GNOME.
All this BS about Qt being “technically superior” to GTK+ is a joke, though. Both APIs have strengths and weaknesses, and indeed, lend themselves to different coding styles. For what Novell needed, it could well be Qt was the better match for. Someone who’s programming in a more esoteric language than C++ might find that GTK+ is a better option. Valid differences, but not really a matter of better or worse.
-Erwos
You addressed everything about oGALAXYo’s post except for the paragraph where he says:
“The next couple of years from my personal view. GNOME will move on the same way it did the past years. Fixing stuff, trashing half finished apps for new approaches, getting new technology into the GNOME plattform which then stays unused or get half implemented (HALF!!!). They will keep developers busy catching up to their permanent changes of deprecating functions and adding new functions.”
Should we expect your lack of reply to mean that he is correct in his assertions, or were you beyond answering such rubbish? Please, do tell.
GNOME has very publically committed to Developer Platform API/ABI compatibility for the lifetime our 2.x releases. So yeah, I’m a bit beyond answering all of oGo’s rubbish. 🙂
There’s always lots to do in our Developer Platform, and it’s getting better all the time. It isn’t as “broken” as oGo would have you believe. KDE are very fortunate to have Trolltech providing great documentation and support for Qt, which puts us at a slight disadvantage in this regard. 😉
We are certainly not ignoring the areas we need to improve on! That would be no fun at all.
Regardless of QT superiority, the problem here is that Novell/SuSE just remains another KDE based distro. A mere distributor of parts, instead of a company who could have *helped* streamline and strengthen GNOME to create an entire “platform”.
Distros suck. We need a platform. And KDE doesn’t help here. KDE is a kosmetic.
Ximian cost $20m, while SuSE cost $210m. It is expected that SuSE would be more central to their strategy than Ximian.
We can disagree, but I don’t think $20m was worth it. They won’t be utilizing Ximian the way they could have, so that’s wasted money. Novell is involved with the GNOME foundation, so I think they had plenty of intentions with Ximian. $20m for the Ximian apps, but without the catering to the expertise behind them isn’t worth it.
Novell can be faulted for being naive though.
oGALAXYo: I think that Novell didn’t really knew WHAT exactly they bought here. My guess is that they initially thought that they bought a Desktop (e.g. they bought GNOME) or that they bought MONO. Or they bought the experts knowledge of those working on MONO. I can’t exactly tell it but guess it only.
Actually, I think they bought it solely for Mono. Novell would love to have a .NET implementation, and while it is nice and all to use an open source project they do not control, it really makes sense for them to buy off Ximian, control Mono (in subtle ways) to fulfil their business plans. I doubt Desktop 2 was really wanted all that much – much of the things done on Ximian’s version of GNOME can be done by almost anyone (key word: most)
Vis: It’s a shame, KDE is just plain ugly as sin.
Ugly as in architecture? Unlikely. Just say, a bunch of key KDE developers want to change the UI of open/save dialog in KDE apps. What do they do? They rewrite kdialog, and after a few weeks of bug crushing, it’s ready to be used. Developers need not to adapt, and if they do, it involves nothing more than a few lines of code.
By comparison, in GTk+ requires the developers to write a new GTK+ component, taking months (as seen recently), and after all that, they have to cohort GTK+ developers to use the new dialogs instead of the old one, which requires a rather lot of adaptation by the developers.
Unless, of course, you mean aesthetics. In that case, it is far easier for KDE to look as good as GNOME (IMHO, a few minor changes and they are already better looking) or as good as OS X, Windows XP, etc. than it is for GNOME to be as good achitecture-wise in comparison with KDE.
Wrawrat: Of course if all the apps were written using qt then it wouldn’t be such an issue, but so many GOOD solid apps like mozilla, gimp, etc are gtk2 ones.
Mozilla is NOT a GTK2 application. Repeat after me, Mozilla is NOT a GTK2 app. It is a XUL application with its own toolkits and APIs, while using GTK+ in the similar manner as they use Win32 and Carbon. In fact, for some time, there was a Qt version of Mozilla.
As for GIMP, it is how GTK was started. So? It isn’t a GNOME application, and even CVS versions of the GTK2-based GIMP (there isn’t a stable out yet for GIMP2) doesn’t follow GNOME HIG anyway. So I don’t see what’s wrong with using GIMP (especially a KDE-fied GIMP using QtGTK – http://dot.kde.org/1073668213/ )
Peter: Today they promise Mono (which definately will lead us into another SCO-like patent desaster) and tomorrow they will hype something else.
Other than GTK#, the fact that Ximian started Mono – I don’t see how Mono is GNOME’s hype. In fact, many within the GNOME community is against the concept of Mono, mostly because of Sun’s involvement in GNOME’s development. While it might be Miguel’s pipedream, don’t expect Mono to replace GTK+ and C, or be a central part of the GNOME desktop.
And BTW, if Microsoft picks a fight with Ximian, they are picking a fight against Novell which would be an industry war. Somehow, i predict the DOJ would have a say in this (Sherman Antitrust Act). Even so, without full .NET compatibility, Mono would ease porting .NET apps from Windows to Linux and provide a kick-ass API.
Roy Batty: Novell buys Trolltech, LGPLs the thing, gets the QT-mono bindings up and running and then sweetness.
Remember, under the Free Qt Foundation pack, Trolltech agrees before it sells off itself or goes bankrupt or stops development of Qt, they would release all the source code under the BSD license.
Peter: That’s just wishful thinking on your side. Name one single technology which has been hyped by Miguel Icaza which did _not_ go into the Gnome core in the past.
Gnumeric, Evolution, etc. (You don’t really understand what GNOME Core means, don’t you?). Besides, Miguel would have a pretty tough fight against members of the GNOME Foundation to get Mono stapled into GNOME.
Jeff Waugh: They’re now shipping various distros, and to my knowledge, using GNOME as their default desktop.
Actually, they ship with the distribution default. In SuSE’s case, that’s KDE. Since Red Hat discountinued Red Hat Linux, HP had drop them mostly except in the server arena (where most of their customers are from anyway.
Anonymous: Yeah, and then comes in M$ and sues the crap out of everyone who ever touched Mono. What a pretty picture…
It’s unlikely that Microsoft would sue Mono anytime soon, especially when Java is still around. Mono can be their poster boy in how open .NET is in comparison with Java (being true or not is not up to debate, it’s beside the point). And when it’s safe to sue Mono without driving away key markets, Mono would already have an important position within the industry if Novell/IBM plays their cards right, or a completely insignificant role in the industry. In the prior, it makes no sense trying to to sue Mono, and in the latter, it would seem like a waste of time and money and it would have the potential of putting Mono back on the map.
And even if Microsoft sues, Mono can change direction and differenciate themselves from .NET. Not binary/source compatible – sure. But it would ease porting, and since Mono runs on Windows (IIRC, the initial development started on Windows), it would act as the role of Carbon within the Linux market – easing application porting from Windows to Linux.
Novells purchase of SuSe was genius. I really thought it was going to be IBM who picked them up, but with the investment from big blue in Novell, it almost amounts to the same thing. I think a purchase of QT by IBM would have some real synergy here and I’d expect them to either do it or help fun a purchase via novell. Like I said, my biggest problem with KDE is that my favorite apps are GTK 2 ones and they look not so hot outside of a GNOME environment. However, I’d agree with whoever posted earlier that maybe just getting down to business and choosing a clear front-runner for business and by extension later, home desktops, is a good thing. Even if it is KDE.
I’m surprised to hear about the environment inside the GNOME camp, I read planetgnome every day but I suppose that doesn’t really give you insight into what goes on.
“Remember, under the Free Qt Foundation pack, Trolltech agrees before it sells off itself or goes bankrupt or stops development of Qt, they would release all the source code under the BSD license. ”
I was unaware of this, it certainly changes things.
GNOME and all GTK+ applications are based on GTK+. KDE and its applications are based on Qt. How, specifically, does Novell aim to integrate both GNOME and KDE into one unified platform? Not to mention the fact that they have extremely disparate views on philosophies regarding software development and engineering.
Novell needs to seriously evaluate which Desktop they intend to focus on and stick with one of them. Selling apples and oranges to the public but calling them pears isn’t going to work. Take a peak into the GNOME environment. Take another peak into the KDE environment. What on earth is similar about the two, or can be unified about them? Absolutely nothing!
Take another look at GNOME and GTK+ applications, and then at Qt applications, their difference is plentifully glaring. For starters, compare Nautilus in GNOME to Konqueror. Do you see any similarities in their behavior? How about Galeon or Epiphany to Konqueror? Do they behave the same? I don’t so.
Combining the best of both worlds is silly when either worlds are fundamentally different and follow alternative ideologies. When I’m presented with a desktop where “Ctrl + q” closes some applications both doesn’t close others, or “Ctrl + Page Up” moves the focus to the next tab in one application but doesn’t on others, what does that speak of usability and consistency on the Linux desktop?
I certainly hope Novell reevaluates their strategy, I’m not buying the unified desktop mantra. That is OS X; that is Windows; that is not Linux/Unix. This is certainly some of the aspects of commercial participation in open source projects that frustrate me. The only thing that has brought GNOME and KDE to the status they garner today has been rivalry and competition. The only thing that brought stagnation to the development of the Xfree project, has been the fact that it had no serious rival and competitor.
Just thought Id back up what Rajan said.
http://kdemyths.urbanlizard.com/viewMyth.php?mythID=13
Hey, you quoted the wrong person. I never said that Moz & the GIMP are GNOME applications. In fact, I said the same thing as you for the GIMP, except in a less elaborated way.
But I just don’t like KDE much. It’s UI is too complex at this point. Hope this changes.
I’m extremely surprised by this anouncement.
Novell also said they are very interested in embedded Linux. Qt/Embedded is a power horse in that market, along with all the other advantages. My guess is that there is a strategic relationship with Trolltech at work here.
Qt simply *is* the platform and a beautiful one. Novell needs a solution now, it is simply not viable to present GTK+ in its current form (documentation, C) to ISVs or bet the farm on a half-finished Mono that has barely been embraced within the GNOME camp itself.
Btw, Qt Designer is open source already.
I’m surprised to hear about the environment inside the GNOME camp, I read planetgnome every day but I suppose that doesn’t really give you insight into what goes on.
Yeah, it does – what you’ve heard from oGo are his own crazed rantings, not a representation of what’s going on in GNOME at all. Planet GNOME is the unedited reality. 🙂
Vis: http://www.kde.org/whatiskde/kdefreeqtfoundation.php
Hey, you quoted the wrong person. I never said that Moz & the GIMP are GNOME applications. In fact, I said the same thing as you for the GIMP, except in a less elaborated way.
Yeah, honest mistake. The original quote was from Vis.
cendrizzi: But I just don’t like KDE much. It’s UI is too complex at this point. Hope this changes.
While I haven’t run CVS versions of KDE (ever), however according to the developers and the changelog, a lot of complexity have been removed – the menus are being rearranged, there’s a new KControl, etc. Besides, technically, it is pretty easy to make all those changes (though enforcing them on every KDE developer would be pretty hard without the support of the core KDE developers).
Using QT as the toolkit doesn’t guarantee that an app will look trashy. There are many sensible, good themes out there for KDE. Granted, GTK’s font rendering is probably a million times better, and GTK has MORE good-looking themes, but let’s hope they can get those things worked out.
What does using QT guarantee? It guarantees that you can have access to all kinds of KDE internals. If there’s one edge that KDE has over Gnome, it’s a developer framework. While Gnome is more of a collection of apps that follow a certain creed, KDE is a stand-alone self-contained desktop environment, with a vast api of features that are fully-capable of competing in the modern desktop arena. It’s also very well-documented.
Where Gnome wipes the floor with KDE is usability. I installed KDE 3.1 on my girlfriend’s machine, and I was throroughly embarrassed to have done so. From what I understand, 3.2 and current CVS are an improvement, but I can’t see this warzone being cleaned up in just two releases. It lacked a decent session manager, there was no notification of hanging processes (so if an app crashed for whatever reason I had to detect it myself and manually run a kill), there were no fewer than THREE text editors, the menus were a convoluted mess, the control panel was a gory battlefield, and konqueror presented me with a way change my mimetype settings as well as my internet security settings from the same preference panel, but that preference panel failed to present me with a way to set my default homepage. Yes, my WEB BROWSER didn’t have a method of changing the HOMEPAGE. Quite frankly, I was unable to believe that this thing was passed of as a legitimate environment. My Gnome installation by comparison is slick. Everything is neat and organized, all panels and buttons HIGified (the HIG really is brilliant), application defaults beautifully set with just enough options to allow some flexibility, and a generally well-done presentation.
Really, if you were to reproduce that beautiful presentation exactly, but instead of Gnome internals you power it with KDE’s framework, you have yourself what appears to be identical (with slight rendering differences) from a user perspective. From a developer perspective, you have a much richer set of tools to work with. Now let’s say while you’re re-implementing Gnome you add a new feature that the old Gnome didn’t have that’s wicked-cool. And there’s something that wasn’t done right before, so you’re going to redesign it and do it right this time. And you pull in some ideas taken from OSX/Longhorn/WhateverOS and some original cool stuff as well. Suddenly you have a veritable utopia, with a decent starting point for developers wishing to develop apps for the system. Also, since it’s Novell, you can (hopefully) be certain that it will integrate mono and have extensive .NET api’s, so there’s no need to retrain existing windows developers….they just need to learn your apis.
Now let’s just hope it’s GPL’d.
… to be very happy about Gnome’s looking, then *work* with Gnome some months (like Novell with Ximian Desktop), then try some hours with KDE and then choose KDE.
This evolution I made was at RH8.0 times, wenn it comes out.
Hmm, maybe everyone must work with Gnome to choose KDE.
Or work with Windows to choose Linux.
Hmm, should flame not so often and understand peoples choice who will choose Gnome for the 1st time.
I made it too and be very happy with KDE 🙂
Is this possibly a mistake? Surely, an announcement like this would have made bigger press than just one site? I realize that Heise is a serious German tech mag, but how can they be the only ones if this is the case?
Supposing it is true, it also must have been decided *extremely* recently, since the iFolder project page (http://forge.novell.com/modules/xfmod/project/?ifolder), says: “iFolder: integrated file sharing in the GNOME, Windows, and OS X desktops”
There’s no mention of KDE there. IBM, not long ago, claimed they were going to start a Linux desktop support program. The slides from that presentation (I apologize, I can’t find a link…) only showed mention of GNOME. I didn’t find a single KDE reference there, yet now suddenly IBM (along with Novell and HP) decide on KDE? It just seems odd that they’d change their minds so quickly.
I guess I’d like to hear Nat Friedman speak on this, since he’s now a Novell executive, too, and very directly in charge of Ximian’s desktop. Surely he would have the inside track on this?
So how does this announcement fit into the whole “best of both worlds” blurb that Chris Stone put out the other day?
It might just mean that they choose KDE as their desktop, create a theme like bluecurve, and the Ximian purchase is just a “talent acquisition”. Hasn’t Ximian put a lot of work in Gtk’ifying Open Office or something? Open Office has to be an integral part of any business desktop. So, if it is just a matter of choosing KDE and doing a theme then I don’t see this having much of a branding effect for the Novell desktop.
For those that think that KDE is butt-ugly, how so? I’m running dropline Gnome right now and it doesn’t look any better than KDE. Seriously, if it’s just a matter of de-cluttering some of the apps then that can easily be done, but bringing Gnome up to the technology of KDE is a major undertaking. Face it, Gnome is just a bunch of random apps that use the Gnome/Gtk+ api and have a HIG. There isn’t much desktop integration there compared to KDE.
The only thing I would get excited about is if IBM or somebody else bought out Trolltech, Qt was opened up, and there were QT bindings for Mono. At that point, you have a kickass gui toolkit that with automagic bindings for any language that targets the CIL. There needs to be something else for KDE other than the defacto C++. Python doesn’t cut it either. It’s just way too slow for many tasks.
I installed KDE 3.1 on my girlfriend’s machine[…]
One does not need to read any further. This is the big problem with people complaining about KDE. They complain that out of the box, there are too many options, too cluttered, unsensible defaults, etc. But forget that SuSE doesn’t ship default KDE – their version of KDE has a lot of spit-and-polish, a lot of sensible defaults.
And this pretty much blows the entire argument in this regard – for distributions, KDE is just as good as GNOME in the usability sector. As seen with Xandros, Lycoris, etc., the distribution can do a lot of changes to KDE to make it better suited for consumers. And corporate users. In the end – how many consumers and corporate users are going to KDE.org, and downloading vanilla KDE for their distributions? Hardly anyone. But how many geeks (that can handle 3 text editors) would be doing that? A lot.
Now let’s just hope it’s GPL’d.
It’s already GPL’ed (or with a more liberal license, like LGPL, BSD, etc.)
There seems to be an aura of real gloom and doom around these developments. Everyone take a step back. This is ONE distribution. They are NOT swallowing up every developer out there. This is NOT the end of Gnome. NOT EVERYONE is going to suddenly dump everything else and start using Novell/Suse. The “community” WILL survive!
In the future, there will be both corporate and community versions of Linux. The corporate Linux world will experience a lot of the evils that we currently associate with companies like Microsoft. So what? Linux has grown and prospered under a Microsoft monopoly so imagine what the community will be able to achieve with access to all the code that will be available to them through HP, Novell, IBM etc. via the GPL.
Is KDE limited to c++, if so you can count me out on every having any interest in it. I quit c++ with MFC stuff.
That’s good that KDE is getting better.
Unfortunatley this all is just a small issue compared to the big one of having two desktops. I really hate that I have to dab in both camps if I want the perfect combination of applications. It’s just a mess and this only shows it further. Hopefully what we’re seeing here is the start of a unification process. I can only imagine how hard this would be but with many power houses pushing for this it could happen. It’s hard to believe these two camps every being friend but it seems there must be some way of unifying these together if not just making it appear that way to the end user (I’m talking about something that goes well beyond bluecurve).
Ximian might have spent lots of time and boatloads of money GTKfying OOo but SUSE did it in a fraction of a time for KDE. Now SUSE is already shipping a KDE-integrated OOo in SUSE 9.1. Check http://kde.openoffice.org/ also.
Really, if you were to reproduce that beautiful presentation exactly, but instead of Gnome internals you power it with KDE’s framework, you have yourself what appears to be identical (with slight rendering differences) from a user perspective. From a developer perspective, you have a much richer set of tools to work with. Now let’s say while you’re re-implementing Gnome you add a new feature that the old Gnome didn’t have that’s wicked-cool. And there’s something that wasn’t done right before, so you’re going to redesign it and do it right this time. And you pull in some ideas taken from OSX/Longhorn/WhateverOS and some original cool stuff as well. Suddenly you have a veritable utopia, with a decent starting point for developers wishing to develop apps for the system. Also, since it’s Novell, you can (hopefully) be certain that it will integrate mono and have extensive .NET api’s, so there’s no need to retrain existing windows developers….they just need to learn your apis.
That’s the thing, you need the KDE framework, documentation, tools with a bit more polish on the apps. With Gnome that’s just not possible. The whole Gnome framework(if you can call it that) is so far behind KDE’s that it would take a huge investment and a bunch of time to get it up to KDE’s level. One of KDE’s advantages is that they’ve been able to concentrate on the higher-level framework above QT with trolltech having full-time employees working on QT. Gnome never has had this luxury.
If Novell is going to do it right then they have to standardize on one desktop. Don’t mess around with this integration crap. It’s just going to look like every other hodge-podge desktop out there and Novell’s branding will mean nothing.
I think the missing ingredient here is IBM. If IBM would make a public commitment to one desktop or the other then I think we could finally have a standard linux desktop. Until there is one desktop to write to, Linux on the desktop will just remain somewhat of a nice that most ISVs aren’t going to touch.
There is not one set of Qt# bindings for Mono.
There are TWO.
Yes you heard right.
One is at qtcsharp.sf.net
The other is in KDE CVS in the kdebindings module.
There are many companies that have people working fulltime on GTK, including Red Hat, Novell, IBM and Sun.
My cousin works for Novell as a programmer and I can say they do almost everything in Java when possible. So this idea of standardizing on KDE as a development platform is interesting. I mean they would be switching to c++ for alot of stuff and I just don’t see this happening (or is there something I’m missing)?
I think it’s clear that there are just too many things not answered by this yet. There is too much good from the Ximian camp that they can’t just ignore and they certainly haven’t made any effort to help the QT# project at this point. Mono development is sure to be mostly done using Monodevelop which is all GTK stuff. Evolution will be their main linux email client (they are making a Groupwise connector) and that is obviously very much a Gnome app. RedCarpet is a big one of their offerings, all gnome. What a mess. Given all this it’s insane to think that they are going to just hop on board with QT. They must really think they have a way of using both.
Good luck with this…
Raynier pointed out the kdebindings module link to me the other day. It was my understanding that this was only for KDE though and not QT though. Am I wrong here? If there are new QT bindings then I’m assuming they’re pretty immature since I tend to follow this stuff and it’s all new to me.
What an uninformed piece of FUD. 🙁
> GTK’s font rendering is probably a million times better
Font rendering is done be the same underlying library.
> Yes, my WEB BROWSER didn’t have a method of changing the HOMEPAGE.
It has and always had.
There are many companies that have people working fulltime on GTK, including Red Hat, Novell, IBM and Sun.
Can they put out some freaking decent documentation then? Every time I look at Gtk+ documentation it looks like someone took a perl script to freaking header files. That is just a non-starter for so many people.
Ximian might have spent lots of time and boatloads of money GTKfying OOo but SUSE did it in a fraction of a time for KDE. Now SUSE is already shipping a KDE-integrated OOo in SUSE 9.1. Check http://kde.openoffice.org/ also.
Actually, what Ximian had done was far more extensive than with SuSE. What SuSE did was to change the font rendering engine and the icons, along with *some* widgets and dialogs. Printing, for example, is still different from other KDE applications, along with most things (i.e. the dialogs). Ximian on the other hand did a pretty significant feat, mostly with hacks. It isn’t a easy task, from what I understand from OOo developers, to change the printing engine completely. Ximian did that. SuSE didn’t.
As for kde.openoffice.org, it is still at the beginnings stage, and most of the proposed/actual changes are comestics (making OOo look like a KDE app), with the exception of making different components of OOo as KPart components. Personally, I hate to use OOo not because of its Windows 9x-like widgets, or its ugly icons, but rather its inconsistency with other applications (i.e. dialogs, wizards (if you can call them that), printing, etc.). Not a deal breaker per say, but definately annoying.
Uhh, I think you’re wrong about QT going BSD if trolltech were to get bought out. http://www.kde.org/whatiskde/kdefreeqtfoundation.php
What’s interesting is that all of this going BSD if trolltech should die all concerns the “QT Free Edition” which only pertains to the X11 windowing system. What’s weird is that I thought the license had changed to straight GPL. The date of the license says 1998. It looks like what would happen is if Trolltech were bought out or died, then the X11 version of QT would go BSD and at that point people would have to implement the windows and other OS parts.
Font rendering is done be the same underlying library.
Yes but there’s definitely a difference between fonts rendered by QT and those rendered by GTK/Pango. You don’t have to believe me: just try it.
Interesting article. I put the chances of it being actual fact at less than 10%. I agree with those who say, that trying to meld Gnome and KDE together is an utter waste of time. They are just too different.
Neither Gnome/GTK+ or KDE/QT are going anywhere anytime soon.
I must say that KDE’s defaults are it’s biggest drawback. Even under SuSe. But I don’t think they’re all that bad. I think double clicking desktop icons is a smarter default for example than single click. I think kicker should be setup different by default. There are some other things. However, it’s not that much of a problem. This stuff is easy to fix. I can fix it all within minutes. I shouldn’t have to, but it’s nice that I can. Hopefully novell will do something about this.
I like gnome but I think KDE is much better (at least for me). I’ve had Ximian on my system side by side with KDE. I was really happy with Ximian gnome, a lot more so than “regular gnome”. But kde was better enough (for me) that I have a strong preference for kde over ximian. My father, who’s mostly a linux newb, *strongly* prefers KDE over Gnome. He’s told me many times that gnome sucks. I don’t feel that gnome sucks but we do strongly agree that KDE looks a lot better than Gnome and is the overall better desktop experience.
I put suse+kde on my girlfriend’s computer. I changed the settings a bit and it’s perfect for her. She’s a newb to computing in general. She was having a hard time with windows XP so I gave suse a shot and now she almost never touches XP. She does most of her computer things in under KDE. Such as homework, instant messaging, web browsing, email, etc. I’ll also mention that kde runs a lot faster on her 1.6 ghz p4 laptop than windows XP. I’m not sure how this is under gnome, but the cups integration with KDE is great. It was a breeze to get the printer up and shared over the network.
I also think KDE looks better than gnome. In spite of some annoying defaults, I find the usability to be superior as well. I find it annoying that I can’t configure things that _really_ annoy me under gnome. Either because gnome doesn’t let me, or I just can’t find how to change it. I don’t like how windows max and minimize in metacity. I don’t want it animated in any way, I just wanted it to happen quickly. I double click a desktop icon after I haven’t used gnome in a while, forgetting there’s no animation letting me know I successfully double clicked an icon. If it’s a slow loading app, I worry something’s wrong because no app has loaded (yet), and the icon has done nothing. KDE gives me a generic little animation that lets me know it worked.
My point is that some people on here are bashing KDE without really taking into consideration that there are other opinions out there. Please, do not try to assume that just because it doesn’t work for you, it must not work for everyone else, or at least the majority.
I will say I’m quite happy with novell’s choice. But I also want gnome to thrive.
additional info, regarding the original article about novell to use qt: that’s just a false message. the “standard” (a very respected austrian newspaper with a groovy it-team) directly asked novell about this. they said, that no decision has been made…
regarding the kde-gnome war:
you “kde guys” (sorry, no disrespect) should realize one thing:
there are a TON of long time linux users out there who have switched from kde to gnome during the last year.
these are mostly people (exactly like me) who used kde 2, because it was way better than gnome 1. and when gnome 2 came out, everybody was “uh, that’s buggy as hell” again. so we used kde 3.0…
but when gnome 2.2 and 2.4 came around, we just switched. it’s easier. it looks better.
the cluttered menus ARE a problem in kde right now. i recently tried mandrake 10 (instead of my fedora) and i was astonished how ugly kde 3.2 looks.
i know there are a lot of problems with gnome either. so i guess everybody should get their act together and actually work on stuff. kde 3.2 as it is now is worse than gnome 2.4. for me. and i am normally a loyal user. i “hated” gnome 18 months ago with a passion. so they actually did something right…
cheers,
marillchen
“additional info, regarding the original article about novell to use qt: that’s just a false message. the “standard” (a very respected austrian newspaper with a groovy it-team) directly asked novell about this. they said, that no decision has been made… ”
So what do you think Heise is? Are they not a respected newspaper? Bit of a strange counter argument, please show me the links of your claim.
hmmm ..
Fab
I don’t know much about GNOME but judging from the recent postings by 2 of the leading developers – Icaza and Pennington – it appears that GNOME is looking for direction – something that has not been an issue for KDE – at least publicly. I think that carries some weight in what desktop the big companies choose to promote, if any.
I don’t think it is a leadership issue. KDE does not have particularly strong leaders (read dictatorial) AFAIK. But I think they have been lucky to have Qt. And visa versa.
If GNOME is in a state of disarray then companies abandoning ship is not going to help. Backing by companies is worth a lot. I think SUN in the long run is aiming for a complete desktop in Java. Why else do this http://groupware.openoffice.org/glow/? And considering that Mono may be nothing but legal troubles or seriously amputated then I think GNOME should switch horse and instead create a desktop in Java. With backing by SUN. It will have no legal troubles, a better license than that of Qt and it will also run on … Windows.
GNOME wouldn’t be on the road to oblivion if they implemented some useful features – such as being able to have new windows open in the center of the screen instead of the top left. However in the quest to make things simplified, many features that were once possible thanks to sawfish are now not possible with metacity. And how about trying to make the code efficient? gnome-terminal can generate 100% CPU usage simply doing a ‘ls -laR’. In eterm, the same command hardly stresses the machine at all.
As Red Hat is really the only company keeping GNOME alive and the fact they are not concentrating on the desktop in any significant way, I would not be suprised to see Red Hat divert resources to other areas. While GNOME will not disappear completely while volunteer developers keep working on it, KDE will move forward at a much greater pace. GNOME will not be able to catch up without the paid developers they will lose.
This is awesome. More “demise of Gnome” posts.
When I read on nat.org or on Miguel’s blog that they are going to start porting all of their cool new stuff over to QT instead of GTK+ or GTK#, I’ll believe what was stated in that article.
Pretty entertaining thread I must say.
“GNOME wouldn’t be on the road to oblivion if they implemented some useful features – such as being able to have new windows open in the center of the screen instead of the top left. However in the quest to make things simplified, many features that were once possible thanks to sawfish are now not possible with metacity.”
Then use Sawfish with GNOME. You aren’t tied to Metacity, you know.
“As Red Hat is really the only company keeping GNOME alive and the fact they are not concentrating on the desktop in any significant way, I would not be suprised to see Red Hat divert resources to other areas. While GNOME will not disappear completely while volunteer developers keep working on it, KDE will move forward at a much greater pace. GNOME will not be able to catch up without the paid developers they will lose.”
Red Hat are adding GNOME hackers currently, not getting rid of them. Heard of Sun, too? In the corporate market, GNOME is looking alot stronger than KDE.
This whole meme about GNOME having inferior “technology” that’s going around is just FUD. If GNOME was really such a mess, it couldn’t be making such amazing strides in usability, accessibility, and applications. Every point-release since 2.0 has surpassed all my expectations, and I’m really looking forward to 2.6. The GNOME desktop is just incredibly fast and clean and the best applications are both GTK2-based and HIGrified: Gnumeric, Abiword, Gimp, Epiphany, Nautilus-cd-burner, Sound-juicer, etc.
If the news appears to be true, it would be fustrating. It’s amazing to see managers thinking in that short time periods. I understand Novell adoption of the ‘single desktop’ wish.
Now assume for a moment, the news would be true and Novells Desktop would be employed on a large number of business desktops in the future using the QT based KDE. It is very likely that other software vendors will thus follow the choice of QT for their Linux apps to appear ‘integrated’. As Novell already figured out: User like that and other software makers know this as well.
One day, a vendor would restrict his commercial success on Linux if he would choose any other toolkit like GTK or wxWidgets. This will be the beginning of QT being a monopoly for commercial Linux desktop apps.
To choose GNOME as a default would imply GTK+ or any of its bindings. And additionally it would also mean a rather consistent desktop for nearly all Linux business deployments in the US, including RedHat, IBM, Novell, HP, and Sun shipping GNOME as the default in the next two or three years.
And since the US is still the major software market, this may lead to GTK+ and its bindings becoming the major toolkit without any threat to pay a monopoly rent one day.
I also do believe that other software vendors are not porting their desktop apps because the ‘default toolkit’ to use is still an open question. If Novell would choose GTK+ or any of its bindings, this would have been a clear signal, and may have convicted some vendors to start porting. It might have been the Linux desktop take-off with more apps -> more users -> more deployments -> more apps.
Please note that this argument is only slighly related to the default desktop, as XFce proves. And the need for a default desktop is only temporal.
All of you that moan about cluttered menus being such a big deal – have you not found the option to configure the menu to your liking?
This whole meme about GNOME having inferior “technology” that’s going around is just FUD
Just because random apps have “usability and have been HIG’fied” has nothing to do with the underlying framework for the platform. Even truthful Gnome developers would tell you that the KDE framework is technically better than Gnome’s. KParts, KIOslaves, etc…not to mention better docs, a better gui toolkit. You can bury your head in the sand and act like Gnome is as technically advanced as KDE or you can see what KDE and other desktops are doing right and try to get them into Gnome.
By the way, I’m typing this from a Dropline Gnome desktop and as far as the user interface is concerned I don’t really have any preference whether its KDE or Gnome. I’m actually starting to enjoy Gnome a bit more than I used to, but the fact of the matter is that as Miguel and Havoc keep on pointing out, if Gnome is going to compete its not such a bad idea to start looking at managed environments with good APIS.
Wasn’t the whole point of having Gnome/GTK under a Free licence that no company could put any restrictions on it? So why does suddenly everybody act as if a business decision by Novell could instantly kill an open source project? Will the community stop using and developing Gnome just because Novell says so?
That’s why me and others keep on saying that the best thing to happen is for Trolltech to be bought and QT to be opened up. Not only do I have a problem with paying for a toolkit if I want to do something commercial, but businesses are crazy if they are going to be constantly guessing what the next per-app/per-developer licensing fee is going to be for their next application. I refuse to touch QT until someone frees it up.
tell me when they will relaese a version w/- the new desktop?
Serious, this is on-topic, I promise. Many people here bashed qt/kde by being c++. what they do not know is:
1. glib is an attempt to bring one of the c++ features to c (classes, vtables, virtual methods) — it works. bug in c++ it’s far easier
2. once I was in this c++ shop, and they said: we don’t do operator overload; it’s dangerous! what if you overload operator+(point&,point&) to subtract the coordinates of the points? and I answered: (a) why in heavens one would do something so braindead? (b) operator overloading has its place; it’s not to be taken lightly, and it’s not to to such brain damage; it’s to overload operator+ to when you would do a method called plus and substitute Vector c = a.plus(b) by Vector c = a + b; which is far more expressive;
3. RAII: Resource acquiring in initialization! No “finally” keyword!!! No “goto cleanup_this; cleanup_this: clean_x(); return” want a file? open it! “fstream f(“gee.txt”, ios::read)”! when it goes out of scope (by way of return, even, or an exception) it will be closed!!! No garbage collection large cycle of hmm-is-this-being-used-yet!
4. Everything else is unnecessary complication.
sorry, had to get this off my chest.
If anyone thinks enterprising young Gnome enthusiasts like Nat and Miguel are just going to turn around and say “Alright let’s forget about Gnome, full steam ahead on KDE.” you’re an idiot. Especially now they have a couple million dollars in their back pockets (this is a wild assumption however I presume that they must have profited somewhat from the sale of Ximian).
“GNOME wouldn’t be on the road to oblivion if they implemented some useful features – such as being able to have new windows open in the center of the screen instead of the top left.”
Yeah, I wish publishers would start printing books where the text starts in the middle of the page rather than the top left. Only joking, this probably should be configurable (or spatial).
It’s also not suitable for people like myself who, while very competant with computers, would rather not have to use a bloody /search engine/ to find the right preference for a given program on the desktop (ie: the Control Center).
ROFL! Google for KDE
Expect to see it in their next release 
~~~~~
Till
Anyone who has been following this stuff will remember an announcement by IBM about six months ago, saying that they were going to be using QT/Embedded for all of their future embedded work.
Now, Novell want to work with IBM, so it makes sense for them to get with the programme. After all, QT/E is a subset with refinements of QT, so it makes sense to standardise on QT and be able to write applications for both the server/desktop and embedded space *in the same breath*. Novell also want to write applications to run across multiple platforms (Linux, Windows, OSX, AIX, Solaris, *BSDs, QNX), which are the QT targets. Again, it makes sense to be able to write applications which can be recompiled for all of those systems.
Imagine a merged programming space, where you write one source, purely depending on QT for anything aside from logic (QT is complete enough to do this… I know, I’ve written a multi-platform data processing server), and it just recompiles… Anywhere. Mainframes to servers, to blades, to desktops, to embedded devices to palmtops to mobile phones. One source, recompiled. It is what Java has failed to do for the last ten years. (I would also point out that there are still ongoing projects to create a mail program in Java, but they haven’t done it in the last five years, and it’s not really getting far now. It’s just too hard.)
Even truthful Gnome developers would tell you that the KDE framework is technically better than Gnome’s.
You’re saying that GNOME developers who don’t agree with you aren’t being truthful. That’s rather presumptuous, wouldn’t you say? I agree that KDE has an edge in some areas and GNOME is better in others, but all in all the tech is equal. In the end, what really matters is the final user experience and GNOME has the definite edge there.
… as Miguel and Havoc keep on pointing out, if Gnome is going to compete its not such a bad idea to start looking at managed environments with good APIS.
Well said.
Who is laughing about the search feature doesn’t have a clue about how the mind works. 🙂
When I think, the idea that appears in my mind is “change the mouse cursor theme, cause I can’t see it”. I don’t think in terms of “setup->look&feel->mouse->cursors”. So I should NOT be forced to know (or remember) whether this option is to be found in peripherals, look& feel, or accessibility. In fact, there is no right place where to put it.
What I really want is to click (or type) the word MOUSE and see every option that is related to the mouse.
The search engine in the kde control center is a great improvement. Shame on you. 🙂
I found this article to be very informative about the problems in gnome.
http://slashdot.org/comments.pl?sid=100742&threshold=0&commentsort=…
Or you could click on the mouse icon in preferences (Gnome and Windows). And pick which pointer theme you want. No sense in making the problem harder than it needs to be.
Yes, oGALAXYo can be “informative”.
The KDE and Gnome argument is all fun and interesting sometimes. But there are a number of people… a rather LARGE number of people who use neither. I for example use FVWM with a number of GTK2 apps. I could just as simply use QT/KDE apps if I felt that they had as nice an interface, furthermore if they supplied me with need to have apps such as GIMP.
The KDE or Gnome argument can go on forever and ever…. and while I can most definitely say I have a preference, it won’t affect me one bit what Novell decides to back. I think it’s a bit necessary for Linux to move into an office market… but I’m not in the office market, and I’d take either Gnome or KDE over Windows XP any day of the week.
Short story is this… for someone like me who uses a bunch of applications that happen to be one tookit, it won’t matter what toolkit it really is so long as it looks the same, feels the same, and runs smooth. I don’t care if it’s QT or GTK, I use GTK now simply because QT applications tend to have poorly designed interfaces… but if KDE fixed that and I had a unified theme engine, I’d be more than happy to begin using a number of QT/KDE applications. Maybe one of the reasons I do prefer GTK apps and in that sense, gnome, is because I realize it’s not as integrated as KDE. I know damn well that I take what I choose out of Gnome as a DE and don’t have issues. Yet the few times in the past I’ve removed QT/KDE apps from a 100% KDE environment they seem to run slower than my grandma on nyquil.
Long story short, there’s still a number of people out there who think “Desktop Environment” is little more than hype for overly elaborate projects that boast integration and unification over clear cut usability, flexibility, and functionality. I for one am a person who thinks that, and I for one will not be using any other Desktop Environment than the one that works, and neither KDE or Gnome as a single solution give me the control and functionality I get with FVWM + Number of GTK2 apps.
I found this article to be very informative about the problems in gnome.
http://slashdot.org/comments.pl?sid=100742&threshold=0&commentsort=…
That post is by no account an accurate representation of the state of GNOME. It was written by a known troll who is on some kind of mission to destroy GNOME.
Last year, Novell was pretty much like any other company hyping linux. Then they bought SUSE and Ximian, and now everyone thinks they are going to kill GNOME.
Here’s the scoop. Ximian was not as actively involved in GNOME anymore. They customized GNOME only, and a lot of their stuff has not really been accepted in CVS. Evolution only in there. Miguel is working on mono mostly it seems now. GTK# is a first class citizen of mono right now anyway.
I could argue that Redhat has a bigger influence on GNOME than anyone else. Owen Taylor, Havoc and many others working on GNOME.
As someone currently testing GNOME 2.6, I can tell you that GNOME has reached levels of polish unprecedented in OSS Software.
The just wroks philosophy is evident too. For example, I configured samba normally, and my network appears in Computer. No having to configure another daemon (lisa0 to do that. Same goes for NFS. One place to configure my proxies.
Epiphany has a good basic download manager. Spatial nautilus is very good too. I am writing a full review anyway, so I will come to that. Many people who think GNOME is no good do not know how to use it.
Sorry, I didn’t know him. OTOH, he seemed to know what he was talking about 🙂
First they buy Ximian giving me the impression we would finally get a proper GNOME uber distro. Now they want to get QT into GNOME – regardless of how they pull this off I’m sure the mismash of apps will look out of place due to vastly different design ideas in KDE and GNOME. Regardless of how much they make KDE and GNOME themes it will still be possible to tell a QT based app from a GTK one.
Personally I would like to see them focus on a single desktop and completely dump the other – to get a platform rather than a desktop. Choice is bad, at least when it comes to the enterprise desktop, it should be standardised. Be that on KDE or GNOME.
marille: there are a TON of long time linux users out there who have switched from kde to gnome during the last year.
If various Internet forums is any indicator, I would say otherwise. Especially after GNOME 2.0 was released and took GNOME’s fame flexibility with it (it is funny, the very same people that condemn KDE for not being as flexible as GNOME turned around and condemn KDE for being more flexible than GNOME). But do you have any substantiated proof? Plus, most Linux users, at least those using it in the past year, are geeks. Definately not your regular blue-collar Novell is targeting.
marille: the cluttered menus ARE a problem in kde right now. i recently tried mandrake 10 (instead of my fedora) and i was astonished how ugly kde 3.2 looks.
Aesthetics plays little role in the IT decissions in the corporate market, and certainly (I hope) doesn’t play much of an importance within Novell’s decission makers. As for cluttered menus, they are eithered fixed in 3.2, being fixed for 3.3, or can be fixed by distributors.
And on aesthetics, in Mandrake 10, it is rather hard to the untrained eye to tell the difference between KDE and GNOME – they are relatively the same. So blame Mandrake. Oh, yeah, I hate Keramik too, but there’s plenty of pretty good themes that are available for KDE, and it is easy for distributors to make their own customized Styles/Window Decoration/Icon set if they want to. Again, not an issue.
Claus: It will have no legal troubles, a better license than that of Qt and it will also run on … Windows.
Funny. Didn’t know that not being open source ( is better than GPL for you. Besides, there’s no talk about merging Mono within GNOME (I’m quite sure Novell would object), or rewriting GNOME to be based on Mono.
opa: Red Hat are adding GNOME hackers currently, not getting rid of them. Heard of Sun, too? In the corporate market, GNOME is looking alot stronger than KDE.
Looks can be deceiving. For example, in Munich – what desktop is used? Hmm, I wonder… Currently, there isn’t a lead contender in the corporate market. GNOME is actually considered only because of Red Hat and Sun support. And now with Novell support for KDE, it is unlikely that GNOME fanboys can claim that the corporate market belongs to them.
clausi: And since the US is still the major software market, this may lead to GTK+ and its bindings becoming the major toolkit without any threat to pay a monopoly rent one day.
It is unlikely in the desktop market there would be anyone other than Microsoft claiming to have hold a monopoly for any time being. Besides, if anyone buys off Trolltech (*hint* *hint* Novell, or IBM), Qt would be completely 100% free (under the BSD license)
stew: Will the community stop using and developing Gnome just because Novell says so?
Actually, Novell after acquiring Ximian have a lot of manouvering room within the GNOME Foundation, and thus, can influence in direction. And if they fail, Novell’s pull out would be disasterous for GNOME, regardless of whether it survives or not.
For curiosity, where do you see clutter in Mandrake 10’s kde menus?
The start menu seems ok to me. The right-click popup menus are ok too…
Also the Kwin menus (e.g. “always on top” has been moved to a submenu).
Much effort has been put into this: the KDE guys have addressed this problem. KDE is now much cleaner.
The control center is not good, I agree. But this is being addressed.
Besides, there’s no talk about merging Mono within GNOME (I’m quite sure Novell would object), or rewriting GNOME to be based on Mono./i]..
[i]And now with Novell support for KDE, it is unlikely that GNOME fanboys can claim that the corporate market belongs to them.
This is all assuming the Heise article is even true, of course. I won’t believe it until Novell says something themselves, because it doesn’t add up or make any sense. The writer could just be some SuSE fanboy with wishful thinking…
Eh. Sorry. Bottom paragraph shouldn’t be italicized.
Most KDE vs. Gnome arguments are based around that fact that KDE is more mature than Gnome. This is not disputed.
However, what I dislike is the constant attacks made by the KDE camp citing every single minor imperfection as a reason why KDE is superior than Gnome.
I like Gnome. I don’t like KDE. KDE people – stop telling me that Gnome is rubbish and KDE is self-evidently better. Show a little respect for what Gnome has accomplished.
> That post is by no account an accurate representation of
> the state of GNOME. It was written by a known troll who is
> on some kind of mission to destroy GNOME.
This is getting too far now. Sorry Eugenia but I think you agree when I reply here. This initial comment was done by me oGALAXYo written as a normal conversation and reply here on OSNews.com some days ago – different Thread (maybe one week ago).
I don’t know where you take the right to say ‘a known troll’. You only say this because you are losing arguments in the points that I have brought up there and thus it’s easier for you to call me a troll rather than answering to the problems I have brought up in that reply. It’s always easier to kill the creditibility of people rather than listening to the problems and have them fixed.
In worst case I am willing to demonstrate EVERYONE these problems with a fresh CVS checkout of GNOME here in my house.
For my history. I joined GNOME around 1999, was working on BALSA email client (look in the Authors file) and worked on GALEON together with the old team where I contributed some parts. I wrote CVSGnome one of the few Building Solutions for GNOME and I wrote Atlantis Webbrowser (also GNOME). I am quite committed to GNOME and I would choose it every day over anything else. But this doesn’t mean I am blind or not looking over the plate and looking and comparing other solutions. The stuff I brought up in that huge comment are things that I can prove all (except the GLADE thing that got fixed meanwhile). Prove me wrong please but do it in a honourable way and not like this.
Actually, Novell after acquiring Ximian have a lot of manouvering room within the GNOME Foundation, and thus, can influence in direction. And if they fail, Novell’s pull out would be disasterous for GNOME, regardless of whether it survives or not.
What would happen? Novell would lay off the people working on GNOME? I’d guess they get hired by some other Linux distributor. (Doesn’t RedHat use Gnome a lot? Sun is also actively contributing to Gnome.)
Probably the most interesting thing about Ximian are Evolution, Connector and Red Carpet Enterprise. I don’t think porting Evolution directly to Qt/KDE is a good option – it may be easier to just rewrite it. Still, both options would consume time and money. If Novell wants quick ROI from buying Ximian, it looks to me like continuing these three programs as they are (Gnome-based) is the best way to go. I didn’t read an explicit “we won’t work on Gnome” from Novell, and I don’t see why they would buy Ximian and throw away most of their work.
This is getting pretty far off topic, but we can’t have people believing oGALAXYo’s lies.
I don’t know where you take the right to say ‘a known troll’.
Trolling: To utter a posting designed to attract predictable responses or flames.
You’ve been spreading lies about the GNOME project for several years now, and if anyone fits the definition of a troll, it is you. You further write:
“I am quite committed to GNOME and I would choose it every day over anything else.”
Which is clearly more lies. Just a few days ago, you wrote: “We in KDE…”
http://www.osnews.com/comment.php?news_id=6429&limit=no#214017
which clearly indicates you are a KDE user. You also wrote the following: “And KDE is definately [sic] much better than GNOME”
http://tinyurl.com/29qj6
Why would you keep using GNOME if KDE is so much more advanced and so many years ahead? Don’t answer, I don’t even want to know. These are just two examples of your trolling, there are hundreds more, here’s another nice one:
“the whole GNOME community is full of fucking cheap retards”
http://www.osnews.com/comment.php?news_id=3108&limit=no#84349
Dear oGALAXYo,
I myself still wonder why you did not quit GNOME and go for KDE or better GNUStep?
Critism must be allowed, even on OSS, but the way you do it, is “unter aller Sau”.
Bottom line is Novell/IBM/HP/Sun are driving the Linux development train now and commercial implementations are what you should expect. Pay as you go, for development tools, cross-patent licensing, and eventually software that is more sophisticated and better integrates into capitalist society. “Free” versions of Linux will remain, just don’t expect them to equal up to the latest commecially available products. You may not like it, but money is what makes the world go round.
It is unlikely in the desktop market there would be anyone other than Microsoft claiming to have hold a monopoly for any time being. Besides, if anyone buys off Trolltech (*hint* *hint* Novell, or IBM), Qt would be completely 100% free (under the BSD license)
Yes, it’s unlikely for a long time. However, using QT for a commercial app means a commitment to QT for a long time and the longer it lasts, the more money you will have invested in the development of your app. The stronger the lock-in.
And it’s highly unlikely that somebody buys Trolltech if that means to loose one of the major assets of Trolltech, namely QT.
IMHO, a better way for Novell would be to investige if the QT dependecy of KDE can be replaced somehow. KDElibs seem to be rather central to KDE, so maybe one could use wxWidgets, gtkmm, Inti or a new binding (pure speculation, no offense meant).
However, that’s just my opinion and it has nothing to do with the ‘quality’ of any desktop or the toolkit but with Linux having a better chance on the desktop.